Letter #85: Paul Graham and Jessica Livingston & Carolynn Levy (2023)
Founder of Artix, Viaweb, and Y Combinator | The Social Radars Podcast
Today’s letter is the transcript of a conversation between Paul Graham (Cofounder of Artix, Viaweb, and YC) and Jessica Livingston (Cofounder of YC) & Carolynn Levy (MD, Chief Legal Officer, and Head of People Ops at YC).
Paul was a starving artist painter in New York before starting Artix, quickly pivoting to Viaweb, selling Viaweb to Yahoo, and then founding Y Combinator. Jessica Livingston is the cofounder of Y Combinator and author of Founders at Work. At YC, she was nicknamed “The Social Radar” as the Partners turned to her to get a final check on whether they should fund the team if they qualified in technical abilities. Carolynn Levy is an MD, the Chief Legal Officer, and the Head of the People Operations team at YC, where she created the SAFE.
This was a particularly fun episode, and I highly recommend you listen to it—Paul, Jessica, and Carolynn’s rapport is beautiful, the conversation is full of laughter, and Paul’s impressions and humor are phenomenal.
I hope you enjoy this conversation as much as I did!
(Transcription and any errors are mine.)
Relevant Links:
Transcript
Jessica Livingston: I'm Jessica Livingston, and Carolynn Levy and I are the social radars. In this podcast, we talk to some of the most successful founders in Silicon Valley about how they did it. Carolynn and I have been working together to help thousands of startups at Y Combinator for almost 20 years. Come be a fly on the wall as we talk to founders and learn their true stories. Today, we're doing something I do every day, talking to Paul Graham, who, as well as being one of the founders of Viaweb and Y Combinator, is also my husband. Paul has been involved with startups since 1995. Before he invented the accelerator, he invented the web app. So there's a lot of information in this episode. But it was also I think, one of the funniest. I hope you enjoy it. So we're really excited to have Paul Graham with us here today.
Carolynn Levy: Yay. Welcome, Paul.
Jessica Livingston: Yeah, Carolynn. As we've been doing these interviews so far, I think every single person we've interviewed has brought up Paul and referenced a huge important impact that PG has had on them, and all of that. So I'm really excited to get you on. So welcome, Paul. Welcome.
Paul Graham: Thank you, here I am.
Jessica Livingston: What I wanted to do, because we... I sort of have an unfair advantage, seeing that I'm your wife. But Carolynn has known you for a long time, too. And I sort of wanted to go back and ask you a lot of questions about early days of things. But where I wanted to get started with something I'm curious about. And I don't know if even Carolynn knows that much about it. But it goes back to what you were doing before Y Combinator. That was, you started a company your own startup in 1995 called Viaweb. And I'd like to talk about the story of Viaweb. But I also want to specifically... and this was the first question I had down was to talk about how you invented web-based software! Because I don't think you talk that much about it.
Paul Graham: Do you mean, what is web based--well, now, web based software seems very obvious. I mean, it's, you use the browser to talk to a program that's not running on your computer. The program is running on some server somewhere, you talk to it through the browser. And in 1995, that was a novel idea. In fact, that was what was novel about our startup. The company was called Viaweb because the software worked via the web. The whole company was named after it.
Jessica Livingston: But when I'm telling a family member of mine that you created the first web-based software, they sort of get, like, a kind of blank look on their face. Like, they can't imagine a time when there wasn't web-based software, I guess, is my point.
Paul Graham: Yeah, well I'll tell you, at the time, it was extremely counterintuitive. Because browsers, I mean, it took me a long time, even though in retrospect it was very obvious, it took me a long time to figure out that we could do something like that. Because at the time, browsers were considered to be just something for consuming stuff on a server, like you wouldn't use that to control software.
Carolynn Levy: Were other people building it at the same time, and you were just first, or was no one building it?
Paul Graham: It wasn't some sort of race, no. I mean, there were so few people building applications for web stuff back in those days. It's not the sort of thing where if we hadn't done it, it would have taken someone five years to figure out how to do it. More like six months. It just sort of happened that we were first because we were building software then. And one of the reasons we were first actually is because we did not know how to write client software. In those days, client software meant Windows software. And we did not know how to write Windows software, because we did not want to know how to write Windows software, we chose Unix software. And so the very first version of Viaweb, the store generator, ran on a server on Unix. I mean, that's what we used as our desktop computers--servers, computers, it was all the same thing. And so we already had a website generator running on Unix, that was the first thing we built. And that was why we thought of making it work as a web app. We thought to ourselves, if you could just figure out a way to control this thing on the server by clicking on links, we'd never have to write Windows software! Wouldn't that be great? We'd never had to learn how to do it.
Jessica Livingston: Didn't you have this idea, like, in a dream or a half awake state or something?
Paul Graham: Yeah, I was waking up one morning and I was half awake. And I thought, wait a minute, could we control the generator? That was what we called it, the software for making websites. Could we control the generator by clicking on links on a web browser? Why not? But it was so counterintuitive when I suggested this idea to Robert, we weren't even sure we could do it. He didn't say, Oh, yeah, obviously, of course, why not? We weren't even sure it would work. We had to like, try and build it first. And it took us a couple of days to build a version. But eventually, we had a version. It sort of worked, no hands, you didn't have to type anything into the command line on the server to make it work. You could control everything, you could like, add an item to your store, delete stuff, publish stuff, generate pages, just by clicking on links on a web browser. It was so non-obvious that that was possible, that we had to build it to see if it was possible.
Carolynn Levy: How long did it take you guys to build it?
Paul Graham: I think it was about two days to build a version to just see if you could control software on the server through a browser by clicking on links. But this version was like unusable, it was just a proof of concept. So to make, I mean--Viaweb, in its time, was very advanced. It was wiziwig. Like, you were making webpages, and as you were making them, they looked like the pages that were actually going to get generated when you were done. So I mean, to make that, like really good looking wiziwig version, actually took us about three months, I think.
Jessica Livingston: So Let's back up a little bit further, because I dove in with Viaweb. And I want to go back to like how you met Robert Morris, and why you wanted to work on a startup.
Paul Graham: I met Robert Morris, when I went to Harvard for grad school. He was an undergrad. And even though he's only one year younger than me, he was actually two years behind, because he got kicked out at one point for putting Harvard on the internet. Harvard was one of the first universities connected to the internet, but by the time Robert got there in the mid 80s, Harvard's internet connection had died of bit decay. It wasn't working. And so Robert basically spent one entire semester getting Harvard kind of reconnected to the internet.
Carolynn Levy: Unbeknownst to Harvard.
Paul Graham: Harvard was a particular computer. Harvard was the name of one deck of computers sitting in the machine room. So like, at one point, my email address was pg@harvard.edu, because I had an account on that machine. So when I say he got Harvard connected, I mean, he got that particular Vax connected to the internet. And it like--he didn't do any work on his classes. And so he like failed a lot of classes, and Harvard has this thing where they kick you out for a year if you fail, to rusticate you, so you can learn how grim the real world is and be more grateful when you come back. So he had done that, he had gone off and had to work for some computer company.
Carolynn Levy: But they weren't grateful that he'd actually fixed their computer. But like, were they like, thanks, and you're still kicked out? Or were they like...?
Paul Graham: Well, I think it was different people. Like the people in the CS department were probably grateful. But this just shows you how little the internet mattered in those days. I mean, it wasn't even called the internet yet. It was ARPANET still. And so the fact that it had died of bit decay showed that no one was using it. Otherwise someone would have complained when it died. That's how little people use the internet back in those days. There's a data point about how little the internet mattered, that Harvard's internet connection died and no one noticed.
Carolynn Levy: And this was what year?
Paul Graham: Ah, let's see when he did that? Well, I showed up there in 1986. I'm not sure. We'd have to ask him. Maybe 1985, something like that.
Jessica Livingston: Okay. My secret plan is to interview Robert Morris on this podcast.
Paul Graham: Hahaha. The Final Frontier.
Jessica Livingston: The final frontier. For sure. We're never gonna get him. But I'm going to try. You show up to get your PhD at Harvard in 1986.
Paul Graham: And he's still an undergrad. And so he still had two years left of being an undergrad. So we got to like hang out and program together and stuff for two years.
Carolynn Levy: How'd you meet, though? Like, how'd you guys come across each other?
Paul Graham: It would have been inevitable, because we were the people who were staying up late at night in Aiken Lab, the computer lab. So I mean, I... we happened to meet at some sort of departmental tea or something like that. I had been told about Robert. I had been told stories about Robert by a guy I knew at Harvard. So he already had a reputation as this super-duper programmer. Super hacker. When Robert first showed up at Harvard as an undergrad, he wanted an account on there, like the real computer in the computer science department, in the graduate department. And they said, "Oh, no, you have to go to the science center and get some crap account on this mainframe." And so Robert sat down in front of a terminal connected to the computer in Aiken Lab and brought it down, then back up again, a single user, with him running it, created himself an account--so, that was that. He didn't ask twice.
Jessica Livingston: Wow. Long time ago.
Paul Graham: So I'd heard stories about Robert's hacking ability.
Jessica Livingston: And then of course, can you touch a little bit on the worm? Because you were with him when that happened, right?
Paul Graham: Yeah, yeah, the worm. So the internet worm--that was, they had just changed the name of ARPANET to the internet. So...
Carolynn Levy: Was he still an undergrad? Or is he at this point also graduated?
Paul Graham: No, he had gone off to Cornell. He was thinking about the ideas that made the worm work before he went to Cornell, I think, um, but he actually did it at Cornell.
Jessica Livingston: Which is your alma mater.
Paul Graham: Yeah. Yeah, yeah, yeah.
Jessica Livingston: So do you remember when that happened, and were you like, Oh, my God?
Paul Graham: The worm, no one would have ever known that the worm existed, except there was a bug in it. That was the problem. The worm itself was absolutely harmless. But there was a bug in the code that controlled the number of copies that would spread to a given computer. And so the computer would get like 100 copies of the worm running on it, back in the day, when having 100 processes running on your computer would be enough to crash it. So the computer would crash simply from the number of copies of the worm it had on it. The worm had no harmful effects. It just copied itself too much.
Carolynn Levy: What was the point of the worm? Like what was the worm going to do of it hadn't done that, had the copy problem?
Paul Graham: Just to see if it would... could be done. Robert was so amused at the thought of making this program that would quietly spread from computer to computer and just sort of sit there running. His dream was like, he would show up at some random CS department and log into their computer and like, check the processes running through their computer, and sure enough, there's this worm, just quietly sitting there, everywhere.
Jessica Livingston: Oh, my God. But it brought down like all these computers on the whole East Coast, and he got busted.
Paul Graham: Not just East Coast, they shut off the whole internet for like two days.
Carolynn Levy: Wow. Two days.
Jessica Livingston: And were you like, did he call you when... or did you just sort of know what was going on?
Paul Graham: Yeah, he called me and told me what had happened.
Carolynn Levy: And you're like, Wow, friend, that sounds bad. Hahaha.
Paul Graham: I mean, it was obvious something was wrong, from where I was.
Jessica Livingston: Okay. So, and we'll talk to him about how he's a convicted felon, hopefully, if we can get him on the show. But, suffice to say, you're good friends with Robert.
Paul Graham: Yeah.
Jessica Livingston: And, so you're getting your PhD, you got your PhD, but then you... I know this about you, you went to art school before you started Viaweb, right?
Paul Graham: Yeah.
Jessica Livingston: So tell us a little bit about the history about why you wanted to start Viaweb and sort of what was going on in your life.
Paul Graham: Well, I used to do consulting to pay the bills while I was trying to be a starving artist. And I would like do some consulting, and then I would run through the money. And then I would have this financial panic, and have no money, and have to try and get some more consulting. And so I just thought to myself, You know what? I am just going to work until I have enough money that I don't have to work again. I'm just going to solve this problem once and for all, instead of doing it piecemeal. That was the idea. So Viaweb was very specifically started to make money. And as you know, like, those are not the founders who make the most money. And we didn't. Viaweb got acquired for $50 million, which was a fairly big deal back in those days. But nowadays, like, you wouldn't even notice that at Y Combinator, would you?
Carolynn Levy: Hahaha.
Paul Graham: Nervous laughter.
Jessica Livingston: Carolynn. You're laughing way too hard.
Carolynn Levy: Well, I was just going to say, I don't know. I don't know how to comment on that. But I want to go back. You were living in Cambridge?
Paul Graham: You mean, when we started Viaweb? No, I was living in New York.
Carolynn Levy: Oh, oh, oh, so you're doing like computer consulting type of jobs for big companies. You're running through all the money buying art supplies and food? Like what are you spending on?
Paul Graham: No, no, rent. Just living.
Carolynn Levy: Just rent, because you're in New York!
Paul Graham: Yeah.
Carolynn Levy: Oh, why'd you live in New York?
Paul Graham: That's where artists were supposed to live.
Carolynn Levy: Oh, that's--oh, okay.
Jessica Livingston: And actually, Viaweb started out as Artix, I know that.
Paul Graham: Yeah. No, actually no, that's not true. Artix was a separate company that we started with the stupidest idea. This is how I learned about bad startup ideas. Because this was how we learned like that this was where the idea of make something people want came from, because that's what we learned from making Artix. Artix was something people didn't want.
Jessica Livingston: What was it?
Paul Graham: Artix was basically software to put art galleries online. Like some gallery online so they could sell their stuff. But that is not how galleries sell stuff. It's not like people look through all images of all the paintings costing $50,000 and then click on add to basket then check out with their credit card. That is not how art is bought and sold, even now. And we thought like, the internet's obviously the future, we were right about that part. And we would go into all these galleries. You know, you ought to be online, and they'd be like, "What's online?" When we demoed them Artix, it was also usually the first time they had ever seen the web.
Jessica Livingston: Oh, my god, wow, no way.
Paul Graham: Can you imagine?
Carolynn Levy: No, what were you thinking?
Paul Graham: We had a hard time convincing people to let us make websites for them for free.
Jessica Livingston: The good news, though, is that a lot of this underlying software could be translated to building online stores, which is what Viaweb did, right?
Paul Graham: Do you remember when I was explaining how we thought of the idea of the web app? It was because we had it half built already? Like we built a generator for websites that ran on the server. So the only conceptual leap we had to have is, oh, could we control the generator by clicking on links, which is not a huge leap. So it was the same with Artix and Viaweb. With Artix, we would make a page for a gallery, and it had all these different artists, and then you'd click on the artist's name, and it had works by that artist. And you could click on one of those, and it would have the individual work with a big picture and text describing it. So it was exactly the same as an online store. So people were just starting to make online stores at this point. This is like, early 1995. And they're all made by hand, by web consultants. But we'd see these online stores and we're like, wait, exactly like the websites we're generating for galleries. And online stores are hot. And websites for galleries are very not hot. So we already know how to build this. The only thing we're missing is the shopping cart. So I basically said, "Robert, ride a shopping cart."
Jessica Livingston: How did you lure Robert into this idea in the first place?
Paul Graham: It was the summer. If it hadn't been the summer, he never would have done it. But it was a combination of arm twisting, him being bored, thinking it might be an interesting idea. This whole web thing was new, it'd be interesting to do some hacking related to the web, and like, I said, "Robert, do me a favor. Help me out with this thing."
Jessica Livingston: Did you say I'll cook you beans and rice the whole time?
Paul Graham: I did used to! I used to make him dinner every night.
Carolynn Levy: That's a good--that's a fair trade.
Paul Graham: Yeah.
Jessica Livingston: So you lure him in, you're working on it with him on the summer, and does it--by the end of the summer, when he has to go back to be in school--because he's getting his PhD at Harvard at this point, right?
Paul Graham: Yes. After he got kicked out at Cornell, he went back to Harvard, and sort of got back into the grad program there, because everybody there knew how great he was and knew he wasn't actually this wicked criminal.
Jessica Livingston: Okay, oh, so hang on. He gets expelled from Cornell because of the internet worm of '88. And then Harvard's like, oh, we'll take you back. We get it.
Paul Graham: He's sort of worked his way in gradually. He started out as a research assistant, and they realized he was genius programmer. And so he became a grad student again. So yes, he was in the grad program at Harvard at that point, which I had just finished a few years before.
Jessica Livingston: Yeah, okay. And so you go back and like, you live in his apartment for the summer to work on this, right?
Paul Graham: Yeah.
Jessica Livingston: And then fall comes around. And is Viaweb sort of taking off?
Paul Graham: It wasn't taking off in the sense that we had users because we hadn't shipped it yet, we hadn't launched the company. But it was taking off in the sense that it was an amazing piece of software, for those days, for 1995, to have this thing where you could build this very good looking, by 1995 standard's website, just by clicking on a few links in the browser, it was just amazingly better than everything else. All the other website builders--it was like a rocketship compared to a horse.
Jessica Livingston: Now, which came first: Julian, your first investor, or Trevor? Because I want to hear the stories of both.
Paul Graham: Julian, I think.
Jessica Livingston: Okay, so tell Carolynn about Julian.
Paul Graham: So Julian was--Julian Weber--was the husband of a painting teacher of mine at Harvard. And he was this super nice guy. He was a business guy, but not a suit. Like, he had been the President of the National Lampoon. [crosstalk]. So he was a business guy, but cool, and a lawyer. So he could do this magical thing of incorporating this, which probably seems... to you, but it seemed astonishing to us, that Julian could just like file some papers and we were a company. So he gave us $10,000, and in return--and did all our legal work, and was on the board, and would tell us really basic things--
Carolynn Levy: Wait, wait, what state were you incorporated in?
Paul Graham: Probably New York, but I can't say for sure.
Carolynn Levy: Okay, all right.
Paul Graham: I mean, he was in New York.
Carolynn Levy: Yeah, that makes sense.
Paul Graham: Well no, maybe Delaware. I don't know. I never--
Jessica Livingston: See? Paul doesn't pay attention to those details.
Carolynn Levy: I know, I know.
Paul Graham: Only you would ask a question like that [crosstalk].
Carolynn Levy: Of course, of course, that was my first... first thought.
Paul Graham: Well, I have no idea. To this day.
Carolynn Levy: Okay, what did he get in exchange for his 10k?
Paul Graham: He got 10% of the company.
Carolynn Levy: Oh, wow.
Paul Graham: I know. Well, honestly, like, this was the deal that YC is based on. Because I remember, when we got bought in the end, like we got bought for $50 million. So even after the dilution, that 10% of the company was worth a lot more than the $10,000. Paid for it. And so I thought to myself, after that, whoa, Julian got a pretty good deal. And then like, half a second later, I thought to myself, but wait, we would have died without Julian many times. And so we got a good deal, too. That's weird--that something could be such a good deal for one side, but also a good deal for the other side. And that was how we knew that YC's deal was, like, that it was needed. We took a deal like that ourselves. In fact, YC was a better deal, because it was slightly more than $10,000 for slightly less than 10%. But that was how we knew that a deal like that was a viable thing, that there was a need for something like that, because we had taken it.
Carolynn Levy: Random aside, did Julian end up doing more investing in startups? Or were you his one and only?
Paul Graham: I think we were--we have to have been close to his one and only. No, I don't think he did any other investing. I mean, there weren't that many startups. This is 1995. Like, there weren't that many startups. And also it wasn't established how to be an angel investor. In those days, you angel invest in some startup, and then the VCs come along and just wipe you out, kick you out of the car, knock you over the head with a pistol and kick you out of the car in some bad neighborhood.
Jessica Livingston: Carolynn, you're again laughing way too hard.
Paul Graham: Because you know they used to do that! I remember when we started YC as we went to our lawyers, and they said don't do this, you'll invest in all these companies as angels, and if the companies are any good, the VCs will just wipe you out of the cap table when they do the series A. VCs used to do shit like that. They're really mad that they can't anymore. I was talking to a prominent VC and somehow we got onto the topic of like, what was wrong in the world now, what was bad about the investing world or something, he said, "The worst thing is all these people who get stock in the company before the company comes to the VCs, and they can't get rid of them."
Carolynn Levy: Wow, that is some savage truth coming from that person. I'm surprised.
Jessica Livingston: Were you talking on Twitter, though? Or was it in person?
Paul Graham: By email.
Carolynn Levy: Oh, you have been writing. Interesting.
Paul Graham: See, I'm not even telling you guys, my closest friends, who it was. I'll tell you afterwards, Jessica.
Jessica Livingston: Okay, I'll be asking about that.
Paul Graham: Got them both by email.
Jessica Livingston: Totally different timeframe back in 1995. And especially also you were in Boston, so not out in Silicon Valley, you're in Cambridge.
Paul Graham: No, nobody knew anything about startups.
Jessica Livingston: So I really want to ask about sort of how Trevor came onto the scene. Because, Trevor Blackwell was one of the founders of Y Combinator. And he's fabulous. But God, if I had a few wishes, one of them might be to have known Trevor in like, 1995, and what he was like.
Paul Graham: Trevor was a colorful character. I mean, I think this was the point where he--you know Trevor gets new ideas about how to run his life, right?
Jessica Livingston: Yeah.
Paul Graham: And this was, the point where we recruited him was the point where his latest new idea for how to run his life was to write everything down on a large stack of index cards. (Laughter)
Jessica Livingston: (Laughter) I know, this is what I want to hear.
Paul Graham: He had this stack of index cards, and whenever he was talking to someone, he would have to like shuffle through it, and get the appropriate index card for whatever we'd be talking about.
Jessica Livingston: Like conversation details? I don't get it. (Laughter)
Paul Graham: No, not conversation ideas, like everything that was going on in his life. He was working on various programming projects, and he had his art, his wife, his family. All these different things were collected on different index cards. I mean we should really ask Trevor, I don't know. I mean, I thought this was... I wasn't fascinated by this. I was floored by it. Like I couldn't remember, once I went up to Robert's office to talk to him and he was talking to Trevor and Trevor is like standing there waving his hands and talking about, "Woah, you know what you should do." And I'm standing behind Trevor, like making gestures to Robert because I wanted... we wanted to go to lunch. I'm like, get this guy out of here. At the time, he just seemed like this hand waving nut job.
Jessica Livingston: But Robert brought him into you. I know the story was something...
Paul Graham: Oh, yeah, yeah. Well, what happened was, after we'd been working on Viaweb for a month, when it was getting to the end of the summer, school was starting, and Robert was like, Okay, I'm done. We've been working on this thing for a month, and it's not done. I remember him saying that. We've been working on this thing for a whole month, and it's not done. And like, we were still working on it like three years later. So I could tell Robert was running out of patience. And I thought, Hmm, we need to recruit some more programmers. Because Robert did all the systems stuff. I can't write that kind of software. I just do front end stuff. So I asked Robert, Who is the best programmer in the graduate program? And Robert said, "Trevor." I said, "Trevor? Trevor is actually a good programmer?" And Robert said, "Despite all appearances, he's actually really smart." So I said, "All right, we'll recruit Trevor." And we did recruit Trevor. And Trevor pulled such a classic Trevor move on his first move. So I go to Trevor and I say, "Look, we're working on this thing. Will you work on it with us?" And Trevor says, "Okay." And I wanted to test him out first, before we actually officially recruited him. So I said, "Write some software for manipulating images" or something like that, I don't remember what it was. And so for like two weeks, I hear nothing from Trevor. And I thought, Oh, this loser, he's completely ineffectual. And then finally, I go and dig him up, and he says, "Here, come and look at this." And he has rewritten our entire software in Smalltalk.
Jessica Livingston: Oh wow.
Paul Graham: Yeah, because at that point, Trevor was a Smalltalk zealot.
Jessica Livingston: I don't know if that's good or bad. I don't know what Smalltalk is...
Paul Graham: Well, Smalltalk is actually a pretty impressive language. But it's definitely the kind of language that... its users are somewhat zealots. And Trevor, Trevor has these fads, right? So this was back when his fad was being a Smalltalk zealot. And so probably partly as a way of learning to be better at Smalltalk, and partly as a way of learning to understand our software, Trevor rewrote the entire thing in Smalltalk, which neither Robert nor I knew how to program in. He didn't really think about that.
Jessica Livingston: God, that is so Trevor.
Paul Graham: Is this not the classic Trevor move?
Carolynn Levy: He didn't do the assignment.
Paul Graham: No.
Carolynn Levy: He did something that was pointless for you guys, because you could never do anything else with it.
Paul Graham: But also very impressive.
Carolynn Levy: But yeah, but yeah, but like, clearly spent all this time, did something kind of hard. Like, okay.
Paul Graham: Yeah! So I basically said, "One, you're hired. Two, we're not using any of this software. Throw it away."
Jessica Livingston: Didn't you always have to sort of, like, make sure you knew what Trevor was doing? Or like, just sort of keep an eye on him? Because he could go rogue a little bit?
Paul Graham: God, yes. Yeah, yeah yeah. Yeah, you had to... it took the combined efforts of me and Robert to sort of sit on Trevor.
Jessica Livingston: Okay. So Trevor's on the team, Robert's on team, you raised some money?
Paul Graham: That fall, we raised money. Yeah.
Jessica Livingston: Okay. And that was hard, right?
Paul Graham: Well, nobody knew like--there were no investors then. These guys that we raised--everybody was a noob, right? We were noobs in raising money, but our investors were also noobs at being investors. They had never invested in any internet companies before. The internet was very hot at the time because of the browser company. Damn, I can't believe I forgot what it was called.
Jessica Livingston: Netscape.
Paul Graham: Yeah. Holy shit. Yes, Netscape. So Netscape, they probably spent a lot of money on PR. So Netscape had drawn a huge amount of attention to the internet, it was sort of a hot thing. And that was why these investors were interested. But they'd certainly never invested in internet startups before. We'd never had an internet startup before. So it was like two people dancing, and neither of them knew how to dance.
Jessica Livingston: But you raised some money. Tell me like, what was one of the craziest stories from Viaweb? I know there were lots. But I want to hear what was one of the craziest ones.
Paul Graham: Let's see... We had a lot of near-death experiences. We were constantly on the verge of dying. So we were at the point where we were just about to get bought by Yahoo, we went to Yahoo to like, do the final--like they had, Yahoo had basically decided to buy us, and they just wanted to have those final meetings where they make sure everyone's a reasonable person, we can work together, and stuff like that. And so we're in this final stage after they've already decided to buy us, and at that point, we were also doing a funding round, because we were out of money. Classic startup stuff. And there was this big investor who had like committed to be in the funding round, and then was gonna back out. And we heard about this while we were having the meetings in California with Yahoo to get bought. And we had to borrow a conference room from Yahoo and say, "Look, dude, we're calling you from Yahoo, where they're like about to buy us in like a week. Don't back out because you are about to make a giant return in like, one week." That was the level of shit we had to deal with.
Jessica Livingston: Oh, my God.
Carolynn Levy: Hey, I have another really lawyerly question. But like, was there a long time between sign and close? Because why didn't you just not take that guy's money so that you... because if you only had a week, could you not pay your utility bill? Like why even bother?
Paul Graham: We never believed that an acquisition was gonna close. This was like the 15th M&A deal. I mean, I'm not making that up. I think I once counted, and we had like, 15 different acquirers claim to want to buy us.
Carolynn Levy: Okay, so you just weren't a believer yet. Even though you're in the final meetings with Yahoo, you're like, this thing could still fall apart. We still need the money.
Paul Graham: It was good practice learning that maxim that deals fall through. I learned it's never a deal till the money's in the bank.
Carolynn Levy: That's true. That's true.
Paul Graham: To this day, I don't know what my net worth is. Because the guy in charge of my finances, I tell him, Don't count the companies that aren't public. Treat them as zero. When he's calculating my net worth, Stripe counts as zero. So I don't actually know the number. All I know is like the liquid number, because I still never believe it until it's liquid.
Carolynn Levy: I agree, actually.
Jessica Livingston: It makes sense. We've seen the hard way, how many things go south at the last minute. So Yahoo, the deal does close, $50 million bucks! You can go back to being an artist if you want, but, you move to California, because you work at Yahoo. And I was just saying, in one of the previous interviews, I was saying, Carolynn, he was there for one year to the day! And I was really emphatic that you quit on the year anniversary of starting there. And I think Paul told me that it was actually maybe a year and a couple days.
Paul Graham: It was probably like a year and three days or something like that. Because you want to leave a margin for error with stuff like that, with like things that have legal implications.
Carolynn Levy: Well, did you hate--Okay, set aside Yahoo for a second. Did you not like living in California? Or was it just like Yahoo and living in California were so inextricably tied, you're just like, I gotta go?
Paul Graham: No, I actually liked some things about California. Yahoo had this office in this office park. Yet one of the big mistakes that Yahoo made is they thought, we're supposed to be like a business. And businesses are boring. Businesses are in these office parks and have all kinds of suits walking around and job titles and shit like that. And so we're going to do all these things because we've got to be a serious business. And now startups have more confidence. They think we started this thing by messing around and being clever hackers and being informal, let's keep that. But back in those days, no, they were telling Yahoo they had to be at business. And so the founders were suppressing their own instincts and making the place sort of boring and business-like. So, it was a grim place.
Jessica Livingston: Didn't you go in like with the same amount of startupy energy? And like after the first week, you're like, Oh, no one else is working like this. Frustrating.
Paul Graham: Yeah, basically. I mean, my whole habit had been working till three in the morning for the past three years. So I just--for a while, I kept doing what I'd been doing. And we just like, tried to write software and stay up late and just do what we wanted and ignore all the rules. Gradually, it sort of wore us down.
Carolynn Levy: Well, aside from that, though, were you ever like, Oh, my god, so many resources, I'm not worried about the power being cut off, I have food, like, was there anything good about being in a place where you weren't resource constrained?
Paul Graham: Well, no. Because we had the resources we needed. I mean, it was good not to have to worry about fundraising. But it's not like, Oh, at Yahoo, I had this great big PR department, or something like that, you know what I mean? So no, it wasn't--the resources didn't make any difference. It was great that I didn't have to worry--I was rich, right? We started the startup to get rich, and here it is, three years later, and I'm rich. [Exhale]. Like, I can look at the bank, and there it is.
Carolynn Levy: Yeah, but were you... but you weren't painting?
Paul Graham: No, I wasn't. Because I was working on...
Carolynn Levy: Because you were working until three in the morning.
Jessica Livingston: Also, I wanted to point out, at Yahoo, you met Geoff Ralston, who was recently the President and CEO of Y Combinator. And you met Tim Brady, who was The COO of Yahoo and then became, later on, a partner at Y Combinator. So a lot of lovely people--
Paul Graham: Tim Brady was my boss!
Jessica Livingston: Oh, I thought lovely Zod was your boss.
Paul Graham: I had two bosses.
Jessica Livingston: Tim and Zod where your bosses. Aww. Was Tim such an awesome boss? Did you just love him?
Paul Graham: Tim was a super nice guy. I mean, he's still a super nice guy.
Carolynn Levy: Yeah, he is.
Paul Graham: He's just a nice guy. Tim is like solid gold.
Carolynn Levy: Agree.
Jessica Livingston: I'm not surprised, knowing you, you couldn't last for that long in a big company. So you left. And then, tell us what you did for the next couple years, before you met me.
Paul Graham: Well, remember, this is sort of reproduced of what happened in my life. Remember, way back in the beginning of this interview, I said I was just going to work until I had enough money and I didn't have to work again? And you've probably practically forgotten that at this stage. Well, so did I. After three years of working on this startup, and then one more year at Yahoo, I'm like, wait a minute, I started this startup so I could get rich because there was something I wanted to do. What was it again? Oh, that's right, painting, painting, painting. So I thought to myself, Well, I did this to paint. I did this to have enough money to work on what I wanted, so, I've got enough money. Let's quit and go work on painting instead. And nobody believed that was what I was doing. When I went to Zod, because at this point, Yahoo options were worth so much. Like this stock had gone up something like 10x since the point where they bought us. So all these options that they'd given us when we went to work there, they bought us, but also gave us some additional options. They were worth a huge amount of money. No one could imagine anybody would leave that amount of money on the table. And so I can remember this conversation I had with Zod, and when he said, "So, what are you gonna do?" And I said, "I'm going to paint." And he said, "You're gonna..." (Laughter). I didn't even realize till he asked me all these questions about what I was going to paint, and I thought, Wow, what a nice guy Zod is, he's got so much interest in what I'm doing, he really cares about me. And I realized years later, I realized it was because he thought I was making it up. And nobody would leave... he thought I was gonna like go start some other internet company and maybe drag a bunch of Yahoo people with me, but I actually was going to paint.
Jessica Livingston: Oh my god. Yeah, you never lie. Like you would never lie about something. I don't think I've ever known you to do that.
Paul Graham: I mean, back in those days, it seemed to Zod an insane decision, I think.
Carolynn Levy: Yeah. But probably to anyone it seemed a little insane. That's funny that he thought you were going to compete, so he's like digging into your crazy painting story. That's hilarious.
Jessica Livingston: So you started painting, but then didn't you move back... you decided to move back to Cambridge and paint?
Paul Graham: Not immediately. No, I tried to paint in California. I thought, All righ, I finally have--all I wanted was basically enough money to live on, and like a quiet place to work. That's basically what I've always wanted in life. And I finally had it. I had money to live on and a place to work. I could buy all the art supplies I wanted. And I thought, Alright, I want to start painting. And now, I always tell founders, when you are done, when you've like sold a company, and you quit the acquirer or whatever, just go and take a vacation. Because you don't realize it, but you are so burned out. Just go and do nothing for six months. Go travel around the world, sit on the beach, go to Rome, whatever. Just don't do any work. And I thought to myself, I have already wasted four years of my career. I need to get to work. So I like... literally the next day, the next day after I quit Yahoo, I started painting. I took like an hour off, or something.
Jessica Livingston: Wow.
Paul Graham: And then started painting. And I didn't know anybody. I didn't... I had hardly any friends there.
Carolynn Levy: What city are you in at this point? Like, Santa Clara? Where, what city are you in? Like, where are you living?
Paul Graham: Up in the Santa Cruz mountains. By myself. Like the Unabomber.
Carolynn Levy: Yeah. Yeah. So like, you didn't move to San Francisco or anything. You're just like, I'm gonna live like the Unabomber in the Santa Cruz Mountains.
Paul Graham: Well, it was pretty. So I just started painting. And I didn't have anyone to talk to about painting, or practically anything. I was just this weird, isolated guy in the mountains.
Carolynn Levy: One quick question. What happened to Robert and Trevor?
Paul Graham: Trevor came with me.
Carolynn Levy: That's what I thought.
Paul Graham: But he didn't quit.
Carolynn Levy: So he's still at Yahoo.
Paul Graham: Because his wife is like, No, you're not leaving and leaving them behind those options buddy. And Robert never even want... Robert was still in grad school. Trevor had finished, actually, Trevor went from a grad student with no money to a rich PhD in like one weekend, because the deal closed about the same time as he graduated. Can you imagine?
Carolynn Levy: Oh, wow. That's crazy.
Paul Graham: But Robert was still in grad school. Plus, there was no way, even if he wasn't, that he would ever have gone out to work for Yahoo.
Jessica Livingston: So Robert stayed in Cambridge, Carolynn. And this is important for Y Combinator's history, which we will get to in a minute. Trevor has moved to join Yahoo, and is living in Mountain View, or Los Altos, or something like that, with his family.
Carolynn Levy: Okay, okay.
Jessica Livingston: Then, Paul, you--tell us why you moved back to Cambridge.
Paul Graham: I didn't move back to Cambridge, I moved back to New York. Because, here's an example of why rent control is a bad idea. I had a rent-controlled apartment, or, for real experts, a rent-stabilized apartment, but, like, nobody knows the distinction between those two. But I had this apartment in New York, where I was living when we started Viaweb. I still had it--with all my stuff in it, because it was so cheap. It was easier to just leave it there and use it as a crash pad in New York for anyone who needed it than actually move. So all this time, I had this like mummified version of my old life as a starving artist sitting in New York behind this multiply locked door. And so after a while of being like, in the Santa Cruz Mountains, like six months or so, I like drove back to New York. And walked into my old life. And it was so weird, because my old life I had been super poor. I used to make like, $20,000 a year. And now, I'm like this multimillionaire, but like, in the same apartment. I could take taxis. I could eat in restaurants and stuff. It was pretty astounding to be in New York.
Carolynn Levy: That's pretty wild, actually.
Paul Graham: To be transported back. It's like, you're poor, you leave. Four years later, you're teleported back into your old life. But now you're rich. It's sort of like some kind of weird novel. And it was kind of exciting for awhile.
Carolynn Levy: You were doing painting in New York at this time... like, you picked back up the art.
Paul Graham: Yeah. But I still didn't really know many people. So when I was looking for where to live, because I had, I mean, my rent-controlled apartment was a really good deal, but it wasn't exactly luxurious. It was kind of grubby, like, that's the other problem with rent-controlled apartments. The landlord has no incentive to make the apartment nice if they can't rent it at market rates. So it was the kind of apartment where, like, the paint on the walls was like an inch thick. Like so many different coats of paint. So I was looking for... I was going to actually buy an apartment, and I was looking for what neighborhood to live in. And I kept thinking, like, where is the Cambridge of New York? Where is the Cambridge of New York? And then I would go back and visit Cambridge, and I'm like, Oh, this is the Cambridge in New York. So, I went back to Cambridge.
Jessica Livingston: So you moved to Cambridge, and you also bought an office in Cambridge.
Paul Graham: When I moved back to Cambridge, I had a specific plan in mind. I was going to build what is now a Replit.
Jessica Livingston: Really?
Carolynn Levy: Oh, that's interesting.
Paul Graham: We had made a web app, and I thought to myself, I'll make this thing, this platform, where anybody can build a web app through their browser. A web app for making web apps. Create software using nothing more than a browser and launch it and have a whole company. It was like AWS and Github, and I was going to make a new programming...you can see where this is leading. It's like too much work. But I went back to Cambridge, and said, "Hey Robert, I've got another great idea! Wanna work on it with me?" And Robert said, "No." He absolutely refused. He's like, I'm not getting--I already burned four years of my life--no, actually, he only burned three working on Viaweb. When he was a postdoc at MIT, and was looking forward to just doing CS research, not having to work on my shit anymore. So he refused to do it. So I said, "All right." And I couldn't get Trevor because he was still working at Yahoo. So I thought, All right, I'll have to do it myself. And I hired a bunch of like young programmers. And we rented an office in a basement in Harvard Square. And for a summer, we tried to build this thing. And then at the end of it, I just thought, this is gonna have to be such a company. I don't want to run another company. That sucked. I didn't like it the first time. So at the end of it, I thought, all right, all. I won't do the company thing. I'll just do the programming language part. Because we were going to make a whole programming language for making web apps which was, of course, going to be a dialect of Lisp.
Carolynn Levy: Not Smalltalk?
Paul Graham: No. Syntax? Are you kidding? So, at the end of it, I thought, Nevermind this whole company thing, we'll just do the programming language part as an open source project. The rest of it had to be a company, because you had to pay for all these servers for people to run their stuff on. But if you just did the programming language part, that didn't have to be. So I thought, All right, we'll just do it as an open source project. And this new dialect of Lisp at the time was called Arc. So the end of the summer, I just bailed. If Robert had been willing to do it with me, I probably would have kept going. Because it would have worked, it would have actually worked. It would have been an enormous amount of work, but it would have worked.
Jessica Livingston: So this is where I remember meeting you. You were working on Arc. And taking flying lessons to get over your fear of flying.
Paul Graham: Yeah, you notice what I said, I went back to New York. I said I drove back to New York. Didn't that sound a little odd?
Carolynn Levy: I just figured you wanted your car. But then like, whoever wants a car in downtown New York? I mean, like, that's--yeah, fear of flying.
Paul Graham: Yeah, I was afraid of flying for a long time. But I managed to fix it using a dramatic but effective method. I learned to hang glide.
Carolynn Levy: Seriously? Oh, so... I thought you were gonna say you became a pilot. But you just... hang glided.
Paul Graham: No, that was phase two. Phase one, I learned to hang glide. And then I took flying lessons. So in the middle of this, there was a point at which I was very comfortable hang gliding, a little bit nervous, like flying in a Cessna 172 with engine just switched off by the instructor and terrified of getting on an airliner. Like that is how irrational fear of flying is.
Carolynn Levy: Yeah. No kidding. Actually, the hang gliding, to me, sounds like such a nonstarter.
Paul Graham: Well, it's much more dangerous.
Carolynn Levy: Yes, exactly. That sounds way horrifying.
Jessica Livingston: I met you, Paul, when you were in the middle of the flying part. And early on in our relationship, you called me, and you're like, "How are you doing?" I said, "Oh, good. How are you?" And you said, "I'm great." And even then, I knew... What is going on? I was like, "Why are you great? I've never heard you say anything besides pretty good." And he said, "Guess where I am?" And I said, "Where?" And he said, "I'm in LaGuardia." He had taken the shuttle from Logan Airport to LaGuardia, and that was the first time he had flown, and you'd never looked back since then, did you?
Paul Graham: For years I hadn't flown. Yeah. I invited Jessica--this is a very romantic story. I invited Jessica to come down and meet me in New York. And at first she thought to herself, can I even do that? Could I just like go to New York now? I said, Come to New York for dinner.
Jessica Livingston: This was at like 4 in the afternoon, Carolynn. He calls me at 4 in the afternoon. And he's like, Come down to New York tonight. Of course, he's staying in the rent-stabilized apartment, which he still has. And I'm like, What? And I--my office, by the way, you could almost see Logan Airport. I was right in downtown. So I was like, Uhh, okay. And I think I just sort of knocked off early, and bought a ticket and flew down to New York that night and we went out to dinner. And then I flew back the next morning!
Carolynn Levy: Seriously, like romance novel type stuff right there. Do you guys remember where you went to dinner?
Jessica Livingston: Yes. It was a Dan--
Paul Graham: Cafe Boulud.
Jessica Livingston: I was gonna say it was the Daniel Boulud place. It was really good.
Paul Graham: Yeah, it was good.
Jessica Livingston: It was very cool.
Carolynn Levy: Go back to something you said, PG, you still have this apartment?
Paul Graham: Not anymore.
Carolynn Levy: Oh, I was gonna say... you meant still as in... in that era of your life. I was like... Wow, holy crap. Okay. Okay.
Jessica Livingston: Yeah, he still had it. And that's where we stayed in New York that night. So we're in Cambridge, Paul. And I know, we've told the story about how Y Combinator got started. But I'm just gonna bring us back to that timeframe. I was working at an investment bank, and then it was after the bubble burst, and they were laying people off, and it was just like, not that fun anymore. And I was interviewing at a VC firm, Carolynn. And they were just sort of taking a while to make a decision. And Paul and I would go out to dinner every night. And Paul would say, "You know what you should do, or this is how the VC industry should change." And you just had all these ideas. And then, so what happened when we were walking down Shepherd St., Paul?
Paul Graham: Was it Shepherd Street? Or was it Walker Street?
Jessica Livingston: Oh, it was Walker Street?
Paul Graham: Well, it was, we were walking home from dinner.
Jessica Livingston: It was Walker street with friendly cat.
Paul Graham: Yeah. So I had also been talking, like just that day, or maybe the day before about Robert and Trevor about whether we could do something together because I missed working with them. And I had just given this talk to the Harvard Computer Society about how to start a startup up.
Jessica Livingston: Which, by the way, just came up, I need to just tell you, Paul, this came up in Steve Huffman's episode, he talked about how he met you at this. So go on to that.
Paul Graham: Yeah, he was there. He was in the audience. And that made me think about investing, because I had meant to be an angel investor, and here it was, all these years later, and I still had never gotten around to it. So I had been thinking about becoming an angel investor because of that talk, and I had been talking to Robert and Trevor about, Could we do something together? And here was Jessica talking about, like, I kept having all these ideas for how they ought to change venture capital when Jessica went to work at this VC fund--like they would have ever listened to her, anyway. And so on the way home, I thought, like, Why don't we just start our own? Nevermind, like, Why do you need these VCs? We'll start our own thing, and you can work for that. I remember the spot where we--like the spot on the sidewalk where we were talking about this, because it only took about a few steps of walking for her to agree.
Jessica Livingston: Oh, I was totally in it. And I was not a startup person. I had always worked for big companies. But I knew this was going to be exciting. Yeah, yeah.
Paul Graham: So that was how YC got started. Although originally we were going to call it Cambridge Seed. That was the first name, for about a week.
Carolynn Levy: That's not terrible.
Paul Graham: No, but do you know what the problem is? I knew from the beginning that everybody would copy us.
Carolynn Levy: Ohhh.
Paul Graham: Boy, did that turn out to be an accurate prediction. And so if we were called Cambridge Seed, then somebody else could say, Well, we're the Cambridge Seed of Silicon Valley. Can you imagine that?
Carolynn Levy: Totally different trajectory right there.
Paul Graham: We wanted to make it clear that like YC was the YC of Silicon Valley, so we decided not to call it Cambridge Seed. But ycombinator.com wasn't taken.
Jessica Livingston: But weren't we out to dinner with Robert and you said, "Well, what do you think about Y Combinator, Y Combinator?" And Robert thought it was cool. I thought it sounded cool, even though I didn't know what the heck a Y Combinator was.
Paul Graham: Yeah, that night at dinner, that was when we decided to call it that.
Jessica Livingston: At the Rialto, right? In the Charles Hotel.
Paul Graham: Yeah.
Carolynn Levy: So Trevor wasn't part of that conversation. But was he still in California at this point? Or had he come back?
Paul Graham: No, he never came back. Because Cambridge wasn't really back for him anyway. He only spent a few years there, he's from Canada. It was just us three in Boston at that point.
Carolynn Levy: And you knew the domain was available?
Paul Graham: Yeah.
Jessica Livingston: You pulled like an all nighter writing up the website, then everything. I think I remember you were up all night or something crazy.
Paul Graham: Not quite up all night, but I was so excited I think I like got up at five in the morning to work on it. We put together YC really quickly and we launched what we then called the Summer Founders Program, which was initially supposed to be just a one off thing. You know how YC has this trick of like, if we want to do something over and over, we say we're just going to try this thing this once, and then secretly we're hoping it's successful, and you keep doing it over and over. Well, this was the first one of these things. Although we really think... did think we were just going to do it once.
Carolynn Levy: This is such an arbitrary question. Random, but like, the orange, when did the orange happen?
Paul Graham: Oh, in the... from the beginning.
Carolynn Levy: From the beginning, okay. And do you just love the color orange? Like, why orange?
Paul Graham: Because it's the opposite of all the colors of all the VCs. Back in those days, when you looked at VC's websites, they were always like, forest green, or navy blue, or something like that. We wanted to appeal to founders, not LPs. Why are they forest green? Because that's the kind of thing LPs like, not because what founders like. They were all focused on LPs. And founders, we just, like, give them a kick with the back of their heel.
Jessica Livingston: But Paul, the walls in your house in Cambridge were painted orange.
Paul Graham: There was one wall that was painted orange. Yes, yes.
Jessica Livingston: You liked orange.
Paul Graham: I know. I like orange, too. It's not like I had to say, Oh, orange would be the right color. I hate it, but we'll use it anyway. No, I like orange.
Jessica Livingston: So Paul created the logo, and then, we launched the site. This...
Paul Graham: The logo's an inside joke, the logo's based on the Viaweb logo. The Viaweb logo is a white V in a red circle. The Y Combinator logo was a white Y on an orange square.
Carolynn Levy: Orange Square. Yeah, yeah, okay.
Paul Graham: So if you ever seen me with like a blue triangle, watch out.
Jessica Livingston: Now, Paul, correct me if I'm wrong, but we did like the website, the interview process, all of that, kind of before we spoke to any lawyers about setting up Y Combinator LLC, right? We sort of got everything going.
Paul Graham: This is going to kill CLevy, but I can't even remember what order this happened in.
Jessica Livingston: Yeah.
Carolynn Levy: I know who your lawyer was, by the way. It was--
Jessica Livingston: It was Goodwin Proctor! Mark Macenka!
Carolynn Levy: Oh my god, yeah, I actually did a deal with... before I ever met you guys, he and I were opposite on a merger. So I knew him before, I maybe knew him before you guys did.
Paul Graham: Probably.
Jessica Livingston: Now Carolynn, this is a tangent, but we met you, I think, the first winter, in winter 2006, or was it the summer of 2006? Because you started representing Sam Altman's company and Square, and stuff like that. Because you were at Wilson Sonsini then.
Carolynn Levy: That's right. And I think it was sometime in 2006. I don't remember what season. I don't remember if you guys had had two batches, or one. I don't think I was even aware that was what your thing was. So I don't remember.
Jessica Livingston: Okay, but we met you pretty early on in the thing. So, Paul, I was literally just listening to the Steve Huffman episode, and he was talking about how they came to interview and we heard the whole story about how we said, "Sorry, the sell food idea doesn't work." And then you called them up and brought them back and cooked up the idea for Reddit and stuff like that. Do you remember that weekend when we interviewed, like that first weekend? Do you remember?
Paul Graham: Oh, very vividly, yeah. I remember that weekend of interviews more vividly than any of the others.
Jessica Livingston: And we sort of thought, Wow, we've got some decent applicants here. And we chose like eight startups to fund, Carolynn. Reddit was in the batch, the Twitch guys were in the batch. Sam Altman, also President and CEO of Y Combinator later on, was in that batch. Like, we were like, this is kind of cool.
Paul Graham: The surprising thing was that we were getting people who had already graduated from college. This was supposed to be a summer program for undergrads, where instead of going to get a job at Microsoft, because no one was getting a job at Google yet. Instead of going to get a job at Microsoft for the summer, start a start up as your summer job! What a crazy idea, right? And then we got all these applications from people who were graduating, or had already graduated. And I remember saying like, Do you understand, like, do you actually want to do this--it's for undergrads?
Carolynn Levy: I have to stop you for a second, because how did you guys think that through? If you're like, start a startup for the summer, and then go back to school? And then what would happen to your investment? Like, what did you imagine would happen to your investment?
Paul Graham: You don't understand. These were supposed to be throwaway startups. The only reason we even did this thing, we thought we were going to do normal angel investing. Where people come to you with deals, and you invest in them or not. The only reason we ever did this batch format was so that we could learn how to be investors. This is a very hackerly approach. But we thought, All right, how are we going to learn how to be investors? We'll just invest in a whole bunch of throwaway startups all at once, and we'll figure out how to be investors. And that turned out to be the model ongoing.
Carolynn Levy: Hang on, when did you guys start calling it a batch? Because that used to throw people a lot, I think...
Paul Graham: Pretty early on.
Jessica Livingston: Pretty early on. It was the only word to describe it.
Paul Graham: Yeah, I think when we had the second batch, we needed a name for the batch... at least by the second batch we would be calling it that. I don't think we ever had any other word besides batch. But also, it's from programming--batch processing. They're--this, like, programming. There's this, you can do things synchronously or asynchronously. So we definitely understood, structurally, what was novel about what we were doing.
Carolynn Levy: I always thought it was from cooking. I learned something just right now. I just assumed it was like cookies. I did not know it was like programming.
Paul Graham: Batch processing.
Carolynn Levy: All right. All right.
Jessica Livingston: So we took over Paul's office, which had just been renovated with like soundproof windows, and it was fabulous. And I moved on in. And that's where we would host our Tuesday night dinners. And we had tables made, with benches, and the...
Paul Graham: A table. There was only one table.
Jessica Livingston: But they'd connect together. They were like...
Paul Graham: Oh, yes, yes, That's right. There were segments of table that we were joined together to make one big long table. Yes. Right.
Jessica Livingston: Yes. So we could break them down when not using them to eat. And then we had the famous benches--tell the story about the famous benches.
Paul Graham: The benches. So, Kay Courteau, the architect, made these benches that were... these very cool looking benches that had these two vertical sort of planks of wood as the legs.
Carolynn Levy: I'm familiar.
Paul Graham: You know about the benches?
Carolynn Levy: Yes.
Paul Graham: The original version of the benches, Kate decided where to put the planks of wood based on like what looked good, which was a little too much towards the middle of the bench. So if you sat on the end of the bench, you would be on the floor.
Jessica Livingston: You'd go flying off.
Paul Graham: So we knew about this problem. We knew about this problem, people sitting on the end of the benches and falling off. And so Kate, when we were having new benches made for the West Coast batches, because we had to duplicate everything, she said, "Do you want to keep the legs where they were?" And I said, "Yeah, lets. What a great idea."
Carolynn Levy: Really?
Paul Graham: Yeah.
Jessica Livingston: He's not kidding, Carolynn. And I have to say, this is like the one disagreement we've ever had related to anything Y Combinator. I thought it was the stupidest thing to do.
Carolynn Levy: Right. Like, let's intentionally injure our people.
Paul Graham: What we really had to watch out for was investors. We used to do Demo Day at YC, right? The founders knew you shouldn't sit on the ends of the benches. But there would be times when we would see an investor walking towards a bench... Don't sit on that!
Jessica Livingston: Okay, moving on from the benches...
Paul Graham: YC had a sense of humor.
Jessica Livingston: We did have a sense of humor. And let's not... I don't want to tell some of the early stories of the senses of humor, because we'll get in trouble. But they were very funny. And most everything I agreed with, it was just the benches that I didn't love. I want to talk about, Paul, I want your opinion, because it's been so long. And I have my own ideas about this, but I want to hear yours. Do you remember some of the early wins that we had with Y Combinator? Because of course, it was a huge experiment. You were basically funding the whole thing. And we couldn't keep going on forever,
Paul Graham: I funded the whole thing for like seven years.
Jessica Livingston: And so like, what were some of the early signs that we had where we said, "You know, this actually, this could be great", or "this is working", or like, "Yes!"?
Paul Graham: Well, halfway through the first batch, we--remember, we thought of these startups as like this throwaway thing. Like summer jobs are throwaway jobs. These startups would be like summer throwaway startups. And really, about halfway into YC, we realized that some of these startups in the first batch could actually be viable startups. And the speakers would have this reaction too, because I mean, I can't blame them, because that was the way we described YC to them, but they would come and visit YC like they were doing some sort of charity thing, helping out people who don't know what they're doing. And they would come away, always the same reactions like, Wow, these founders are really good, these companies are really good. These are real. We didn't think they'd be real. And so it was--we realized that first, of course, since we had the most data, so I mean, a couple of weeks in, we realized these companies could actually be real. And so by the end of the batch, for sure, we knew that this was the way we were going to do all the investing going forward. This thing we had initially done just to learn to be investors. This was the actual thing.
Carolynn Levy: Was that your aha moment? Like, we should do this in the winter, too? Or were you still planning only to do the next year's summer?
Paul Graham: I think we decided we were going to do a winter batch pretty early on during the summer batch. The big decision was to do it in California.
Jessica Livingston: That was the big decision. That I remember. Because I spent a lot of time like setting up the office in Cambridge and getting all the stuff we needed. Paul would cook dinners every week. So we needed all these crock pots and rice cookers and 20 bowls and sets of... So I set all this stuff up, and I was like just breathing a sigh of relief. And Paul's like, we need to go do this in California. And I'm like, Wha--. Like, I remember just being like, Noo... we have... and like, we had to find a place to live, and all of that. And we asked Trevor, like, "Hey, Trevor, can we use like one of the rooms in your robot warehouse?" He had the Pioneer Way office for Anybot, his robot company.
Paul Graham: If we'd had time, YC would have been in Berkeley. Because back then, we thought Berkeley was the place. Because we liked Cambridge, and Berkeley was the Cambridge of the Bay Area.
Carolynn Levy: You didn't think Stanford was the Cambridge of the Bay Area? That's interesting. I think Stanford thinks it's the Cambridge of the Bay Area, don't you?
Paul Graham: Stanford might be the Harvard of the Bay Area, but Palo Alto is not the Cambridge of the Bay Area. [Chuckles].
Carolynn Levy: Oh, I agree.
Paul Graham: Berkeley is the Cambridge.
Carolynn Levy: Yeah, yeah, yeah, okay.
Jessica Livingston: So we moved, like, with zero notice. We just had to replicate this whole thing. And frankly, there was some light renovations that had to happen for us to move in there. We had to rebuild all the tables, and the benches, and get all this stuff. And we did this sort of undercover, light, renovation, if you will, under the darkness of night.
Paul Graham: We paid to have the builders working at night.
Jessica Livingston: Oh, god. It was a bit stressful.
Paul Graham: We didn't have any permits.
Jessica Livingston: Yeah, yeah.
Carolynn Levy: Okay, yeah.
Jessica Livingston: It's been long enough that I'm not worried about it, but...
Carolynn Levy: I was just gonna say, maybe the time has, enough time has passed that this isn't that controversial, but yeah.
Paul Graham: Almost 20 years. 17 years.
Jessica Livingston: But it was stressful, and we literally, like the first dinner, the like paint was... we always joke, the paint was still drying on the walls. Like, we were like, don't touch the walls, because the paint's still wet.
Paul Graham: Yeah, we had to flush the painters out in order to start the first dinner.
Jessica Livingston: Yeah. But we had, what, 10 startups, or 12 startups, in that batch, Paul?
Paul Graham: Something like that. There were 8 in the first, there were probably like 10 or 11 in the second.
Jessica Livingston: Yeah. And it was really good. And we had Demo Day, the demo day, I remember, was better attended than the one in Boston.
Paul Graham: California demo days were always so much better than Boston demo days. In fact, later, we felt so bad for the Boston startups that we would fly them out to do a second demo day in California. And that was when Drew Houston--that was when Dropbox got stolen by California investors. Like there were these Boston investors who, like, looked upon Drew as their protege. They had been following his progress the way VCs are always gonna do.
Carolynn Levy: Did you guys invite people? Or was it... or did you get inbound for demo?
Jessica Livingston: Hahahahaha. I love that you think we got inbound! Oh my god, Carolynn.
Carolynn Levy: Okay, let me tell you the origin of the question. You're like, Demo Day was always better attended. But I'm like, in the very early days, like the first one, was it just a small... but then, like, I'm trying to figure out how did it grow unless you guys were inviting more people.
Paul Graham: We were inviting people! That was the only way we ever got anybody there.
Carolynn Levy: Okay. Okay. Okay.
Jessica Livingston: PG didn't know anybody there! But we knew--
Carolynn Levy: So how was it bigger?
Jessica Livingston: But we knew... 20 people. And we knew...
Paul Graham: The first demo day in Cambridge, I mean, there was probably only one person whose actual job title was investor in the audience. The rest were all just like, random rich people I could get to fill seats so the startups could feel like they had an audience to present to. I think there were only like, 15 people in the audience at the first demo day.
Jessica Livingston: So Paul, what were some other early wins, like in the first few years that you remember, signs that you interpreted as like, Wow, this could be good?
Paul Graham: Well, when Reddit got bought... when Reddit got bought by Condé Nast. That was the first exit. I mean, and in retrospect, it was comically small. But it seemed pretty significant to us at the time.
Jessica Livingston: Right. And I actually was telling Steve in the episode, I was secretly very pleased by that acquisition, mostly because we got noticed by the press after that. Do you remember?
Paul Graham: And funding rounds, too. Some of these companies were going off, and like, I remember like, when Sam, who was in the first batch, raised a Series A from Sequoia, like, THE blue chip VC firm. We couldn't believe it. We were invited, at one point, maybe that winter, maybe it was a year later, we were invited to Sequoia's annual holiday dinner, or something like that. And we're like, Sequoia invited US to their event! We were like high fiving each other. We're big time now, baby.
Jessica Livingston: Oh, it was so big time. So big time. Like, that was probably, that room was filled with so many movers and shakers. One of whom was Dalton Caldwell, doing Imeem.
Carolynn Levy: Ohh, no kidding.
Jessica Livingston: Do you remember, Paul, when we first met Dalton at the Sequoia thing?
Paul Graham: That was where we first met him? Interesting...
Jessica Livingston: I'm pretty sure, I'm pretty sure. Anyway, Paul and I are not naturally good schmoozers. You have to admit that, right, Paul?
Paul Graham: [Chuckles]. Oh, I'm not. You are, sort of.
Jessica Livingston: So we didn't, like, do a lot of--well, we didn't like do a lot of schmoozing, so it was hard for us to... we met a lot of investors through our portfolio companies that investors would invest in them and then want to come to Demo Day. So that's sort of how it grew.
Paul Graham: That was how we met you, CLevy.
Jessica Livingston: Yeah.
Carolynn Levy: I thought you met me--well yeah, I was gonna say... it's wasn't though... yeah, it was through Sam, because I found your common stock purchase agreement, and I was like, What is this piece of crap? I reached out, and I'm like... and Sam's like, Let me connect you to Jessica, and I was like, Fine.
Paul Graham: Was our common stock purchase agreement not good? This is news to me!
Carolynn Levy: It was fine. I just--it was not... it was very east coast flavored, and I was like, I don't even get what's going on here, and I... anyway... it's...
Paul Graham: Well, needless to say, I never read it.
Carolynn Levy: It got all fixed.
Jessica Livingston: That's because I handled all the legal stuff. Can you imagine me doing all legal and all finance, Carolynn, all finance. I like, hired lawyers, and I hired like, we had this like accountant doing our taxes and stuff. But it was, early on, a lot of work.
Paul Graham: And air conditioner delivery.
Jessica Livingston: And air conditioning delivery, in the very first batch. What other crazy moments were there in the early days, Paul, that you remember? What stands out?
Paul Graham: I don't have stuff indexed under crazy moments.
Carolynn Levy: You don't have index cards that you can just whip out right now and start reading--? (Laughter)
Jessica Livingston: Crazy moments at YC. Just... let me just find my index card.
Paul Graham: There was a constant stream of crazy moments. You remember, like our, the official tagline of YC is like, Make something people want, and the unofficial one is like, You can't make this shit up. Because like...
Carolynn Levy: That's a file. That's a file.
Jessica Livingston: Carolynn currently maintains that file. She maintains the file you can't make this shit up.
Carolynn Levy: Yep, yep, yep.
Paul Graham: Because every day... after there were enough startups, every day there was some freaky disaster. It's like, what's happening today? Ahh, yes. This.
Jessica Livingston: Okay. Well, let me be more specific with a couple things then. Because like, one thing I wanted to ask you about is one of your superpowers, which is just, taking an idea and expanding it like 100 times bigger for founders. Where do you think that comes from?
Paul Graham: I don't have to do the work. [Chuckles].
Jessica Livingston: Well, I think a lot of people--like I know, I'm very envious, I would love to be able to do that. And I can't. Most people can't do that. Where do you think it comes from?
Paul Graham: You know, it might be related to writing essays. Because when you write an essay about something, you have to really completely understand it. And so once you understand it, then you could see where it extends to get bigger. So it was... it's just talking to them and like really, really understanding the idea. And then you can see Oh, you can stretch this bit out and that bit it out, and it becomes triangle or something like that.
Carolynn Levy: That's actually a good segue, though, because we've talked about painting, and we've obviously talked about coding. What about writing? Like, when did that start getting super serious for you?
Paul Graham: About 2000. So between leaving Yahoo and starting YC. In fact, YC grew out of an essay. I would give talks as a way of writing essays. Because if you have to give a talk, you're gonna get up in front of all these people, you better have something to tell them. And so that would inspire me to write essays. And so the talk to the undergrad Harvard Computer Society was an essay called 'How to start a startup' that you can find online. That was the essay that YC grew out of. So a lot of people think that now, since I'm writing essays, I'm like some guy who got rich from startups and now I'm like, sitting around pontificating, but actually, the essays preceded investing. In fact, YC grew out of one of the essays. So I'm just... I just sort of went back to what I was doing.
Jessica Livingston: I remember when I read Why Nerds Are Unpopular. I was like, Wow, this guy's pretty interesting. That was when you asked me out. I found that online. That was one of his first essays, Carolynn. Have you ever read why nerds are unpopular?
Carolynn Levy: Absolutely. And I had my kids read it too, because it talks about high school, if I remember... I haven't read it in a while, but like, and I was like, this is--so my kids were younger, but I was like, this is a really interesting look at high school, which is in your future, guys. So like, internalize this. It's interesting. And I don't remember what they thought of it. But anyway, yes, I definitely have read that essay.
Paul Graham: Mm. Did you read that when I was asking you out?
Jessica Livingston: Yeah. Because of course I had to google you!
Carolynn Levy: She's doing her homework! She wasn't just gonna randomly date some guy she hadn't like, totally dug into.
Jessica Livingston: And I have to say, I don't think I really had much practice in dating nerds. And so, it was good to read. I appreciated it though. And I've always--I do think I'm your biggest fan and I'm glad you're writing now. Because I love reading your essays. And your essays, in fact, are the way we marketed Y Combinator early on. We didn't really do much press outreach, we certainly didn't do anything like paid advertisements or anything like that. It was your essays.
Paul Graham: Well, it wasn't, but they were--it was the essays I'd already written. If you say, when you say we used the essays to market Y Combinator, it's like I'll write an essay and get attention. No, it was--that was where the deal flow came from. Because most investors, if you start some investment firm, you're like, Okay, where am I gonna get a deal flow? But YC, we already--we had a sort of initial source of deal flow in the people who read those essays. So the essays were important to YC for that reason, too. YC grew out of one, and it was the source of the founders.
Jessica Livingston: But you also kept writing these amazing and insightful essays about startups, about venture capitalists, about startup equals growth, some of the canonical--do things that don't scale. These essays that people to this day are like, Paul, this essay changed my life, kind of thing.
Paul Graham: You know what, though? It was not so-called content marketing. I wasn't--like, I was writing--the reason I was writing about startups is because when you're writing essays, you always write about whatever you happen to be thinking about. And so after we started YC, I was thinking about startups all the time. In fact, more than I wanted to. And so I was writing essays about startups all the time.
Jessica Livingston: That's true. I didn't mean to say that you wrote it as a way of marketing. I didn't mean that. I just meant to tell Carolynn, like, it was part of our organic growth.
Paul Graham: It worked out that way. You know, Carolynn, you know how we're always looking for like a startup that grows organically out of the founders lives? Boy, did YC grow organically out of me and Jessica's lives. In so many different ways.
Carolynn Levy: Yeah.
Jessica Livingston: I know we have to wrap up pretty soon. I have just two things that I want to ask you about. And I feel like maybe someday we'll do a part two for Paul Graham, because we didn't scratch the Y Combinator surface that deeply. But...
Carolynn Levy: We did not.
Jessica Livingston: I want to ask you... Well, first I want to ask you about when you retired in 2014. I want you to tell the story about that, because Y Combinator was like, on a high. So why'd you leave?
Paul Graham: Yeah. Well, because I hadn't actually meant to be an investor, remember? I was like writing essays and software. And then we thought, Okay, we'll start a VC thing as well. And this thing that we started as well, it was meant to be sort of a part time job, ended up becoming more than a full time job. It was like gradually taking over my whole life. And I mean, I could see it, I could have kept doing this until I died, because YC is still going, after all these years, right? There's no signs it's gonna stop going. So at some point, I'm either gonna do this and nothing else for the rest of my life, or, if there's anything else I want to do, I have to leave at some point. But it was Robert, really... Robert, at one point--Robert NEVER volunteers advice. Like never. In fact, even trying to get him... trying to get advice out of him is like pulling teeth. He was once reviewed in some undergraduate guide to like, professors and grad students or something at Harvard, and they said, "Getting him to talk is like pulling teeth." That's what Robert's like. It's gonna be... It's gonna be real interesting if you ever manage to get him on an interview, you're gonna have to deal with the monosyllabic answer. But you both have teenagers. So you have some practice too with the monosyllabic answer.
Carolynn Levy: This is true. Teenage boys, specifically.
Paul Graham: Yeah. So we'll be interested to see who wins. Irresistible force or immovable object?
Jessica Livingston: That's true. Yeah. So Robert, say what Robert said to you.
Paul Graham: Robert would come out for interviews to California. So Robert was there visiting for interviews. And it was always fun when he would come, because it wasn't just for interviews. It was like, we get to hang out with Robert. And so we were walking down to the center of Palo Alto, and at one point, he stopped and said, "You know, you should make sure YC isn't the last cool thing you do." And I couldn't even understand what he meant. But it took me like months to realize, but I sent him an email. And I said, "Like, when you said that, you meant I'm wasting my time, didn't you?" He said, "Yeah, basically." He said, "You've already proved your point. Like it works. Why are you still doing it?" Which is not the normal model of people working on companies. When it works, that's we keep doing it! But Robert has a different view of the world. And so, I sort of thought to myself, Maybe he's right. And then my mother got sick. And I used to have to fly up to visit her every weekend. So on one of these flights, I was thinking, All right, I remember looking out the window of the plane and thinking, Okay, I'm gonna find somebody else to run YC. So we tried to recruit Sam. Because Sam was then becoming available. I think that was part of it. It took me a long time to recruit him, though. Like over a year.
Jessica Livingston: Yeah. That was in the works for a while.
Paul Graham: But the Fall of 2013, he finally agreed to do it. We went to talk to you, CLevy, like, what do we have to do to do this? And you're like, Oh, don't worry, it won't be that hard.
Carolynn Levy: Oh, yeah. It won't be that hard.
Paul Graham: But for a while, you were the only person who knew.
Jessica Livingston: Yeah. I have one last area that I just want to quickly touch on, because one of Carolynn and my hopes is that we're gonna get the current man behind Hacker News to come on, who I--will remain nameless at this at this time. But you started Hacker News. And does it bring back a little trauma?
Paul Graham: Yeah. Oh, that was the majority of the pain of doing YC, was Hacker News. Yeah, I always tell people, my advice about forums is Don't start a forum. It's so grim.
Jessica Livingston: But it was so good, though.
Paul Graham: It was good for YC, it was not good for me. But like more than 50% of the stress was Hacker News. I mean, if all I had to do was read applications, pick startups, and then help them, my life would have been so easy.
Jessica Livingston: Let's see. Carolynn, is there anything you wanted to ask Paul?
Carolynn Levy: Well, I kind of feel like we do need part two. So maybe we bring Paul Graham back sometime in the future and ask more questions.
Jessica Livingston: I think I can just--Carolynn, I'll put it on the calendar and just tell him he's going to do it.
Carolynn Levy: Okay, that sounds good. Oh, I do have one.
Paul Graham: By the way, that's how everything in my life works.
Carolynn Levy: I have one final question. Which is that, right now, do you paint and write in equal, or one a lot more than the other? Or how do you...?
Paul Graham: No, I just write.
Carolynn Levy: You just--and what happened to painting?
Paul Graham: I did try it, with, in fact, do you remember how when I quit Yahoo, like the next day I started painting? Well, when I quit Y Combinator, the next day I started painting, because by that point, I couldn't go off on vacation anyway, because I had little kids. So I did. I like-- and in fact, I went up the hill, to that house in the Santa Cruz Mountains, the same place, even...
Carolynn Levy: You Ted Kaczynski'd it again.
Paul Graham: Yeah. But I will--it worked better this time. Because I knew some people now, in California, partly...
Jessica Livingston: And he'd come home to us at night, after painting.
Paul Graham: Yeah. Like going off to a job. But I painted for almost a year after I quit YC. And then I got to a point, I was in the middle of a painting, and it felt like work. It felt like such a schlep. Like usually I was excited to see how the thing was going to turn out, and I'm like, [long exhale], I don't know what it was. But it--I just, I mean, I might do it again. I haven't completely given up. But I do think I'm actually a lot better at writing than painting.
Jessica Livingston: I think you're a great painter.
Paul Graham: I can tell--when I go to art galleries, I look at the paintings and I think, Wow, this guy is amazing. I don't even know how he does it. And when I look at writing, even by the most famous writers, I think, Uhh, I don't know, I wouldn't have used that sentence.
Carolynn Levy: Oh, interesting, interesting.
Jessica Livingston: I think he's a great painter, Carolynn. But I mean, honestly, I think you're just such an amazing writer that that's what you need to be doing. All right, Paul, how about I make you a cup of tea, we'll meet up in the kitchen over a cup of tea.
Carolynn Levy: I think you guys get to have a cocktail now, right, given the time? I mean, I don't, but you guys could.
Jessica Livingston: Actually, we might have a glass of wine instead. But this is a lot of fun. It was really fun walking down memory lane here. So thank you for coming.
Paul Graham: It's certainly a lot more fun talking to you guys than talking to normal journalists.
Carolynn Levy: Aww.
Jessica Livingston: Aww. Damning with faint praise.
Paul Graham: Okay! It was a lot of fun talking to you...
Jessica Livingston: So it was a lot of fun. And yeah, well, can't wait for this to run, and we'll plan part two, Carolynn. Okay, bye Paul.
Paul Graham: Okay.
Carolynn Levy: Bye Paul.
Jessica Livingston: Thank you.
Paul Graham: Bye CLevy.
Carolynn Levy: That was a very--I mean, that was so much fun. Look what time it is.
Jessica Livingston: I'm so sorry that I let it go that long.
Carolynn Levy: No, no, no. Oh my gosh, no, I don't mean to say that as like a critique. I mean, like, that was so much fun, and we had so much to talk about that time like went by really, really fast. That was great.
Jessica Livingston: It did, and I think maybe someday we'll do part two.
Carolynn Levy: Doing a part two with him would be a lot of fun, especially because people will have heard the founders' stories more by that time. And then having Paul tell the... you do a lot of the YC side of it, but also having Paul tell the YC side of some of the founder stories will be, I think, a good like loop.
Jessica Livingston: Yeah. We'll come back to him after we do some more YC founder interviews. And we'll get him back and get his side of the story, because I agree, that would be a good take. But awesome. Well, that was a lot of fun. Nice to see you, Carolynn. And we'll see you later. Okay, see you next week.
Carolynn Levy: Bye!
Jessica Livingston: Bye!
Wrap-up
If you’ve got any thoughts, questions, or feedback, please drop me a line - I would love to chat! You can find me on twitter at @kevg1412 or my email at kevin@12mv2.com.
If you're a fan of business or technology in general, please check out some of my other projects!
Speedwell Research — Comprehensive research on great public companies including Copart, Constellation Software, Floor & Decor, Meta, RH, interesting new frameworks like the Consumer’s Hierarchy of Preferences (Part 1, Part 2, Part 3), and much more.
Cloud Valley — Easy to read, in-depth biographies that explore the defining moments, investments, and life decisions of investing, business, and tech legends like Dan Loeb, Bob Iger, Steve Jurvetson, and Cyan Banister.
DJY Research — Comprehensive research on publicly-traded Asian companies like Alibaba, Tencent, Nintendo, Sea Limited (FREE SAMPLE), Coupang (FREE SAMPLE), and more.
Compilations — “A national treasure — for every country.”
Memos — A selection of some of my favorite investor memos.
Bookshelves — Your favorite investors’/operators’ favorite books.