February 2021Before college the two main things I worked on, outside of school,were writing and programming. I didn't write essays. I wrote whatbeginning writers were supposed to write then, and probably stillare: short stories. My stories were awful. They had hardly any plot,just characters with strong feelings, which I imagined made themdeep.The first programs I tried writing were on the IBM 1401 that ourschool district used for what was then called "data processing."This was in 9th grade, so I was 13 or 14. The school district's1401 happened to be in the basement of our junior high school, andmy friend Rich Draves and I got permission to use it. It was likea mini Bond villain's lair down there, with all these alien-lookingmachines — CPU, disk drives, printer, card reader — sitting upon a raised floor under bright fluorescent lights.The language we used was an early version of Fortran. You had totype programs on punch cards, then stack them in the card readerand press a button to load the program into memory and run it. Theresult would ordinarily be to print something on the spectacularlyloud printer.I was puzzled by the 1401. I couldn't figure out what to do withit. And in retrospect there's not much I could have done with it.The only form of input to programs was data stored on punched cards,and I didn't have any data stored on punched cards. The only otheroption was to do things that didn't rely on any input, like calculateapproximations of pi, but I didn't know enough math to do anythinginteresting of that type. So I'm not surprised I can't remember anyprograms I wrote, because they can't have done much. My clearestmemory is of the moment I learned it was possible for programs notto terminate, when one of mine didn't. On a machine withouttime-sharing, this was a social as well as a technical error, asthe data center manager's expression made clear.With microcomputers, everything changed. Now you could have acomputer sitting right in front of you, on a desk, that could respondto your keystrokes as it was running instead of just churning througha stack of punch cards and then stopping. [1]The first of my friends to get a microcomputer built it himself.It was sold as a kit by Heathkit. I remember vividly how impressedand envious I felt watching him sitting in front of it, typingprograms right into the computer.Computers were expensive in those days and it took me years ofnagging before I convinced my father to buy one, a TRS-80, in about1980. The gold standard then was the Apple II, but a TRS-80 wasgood enough. This was when I really started programming. I wrotesimple games, a program to predict how high my model rockets wouldfly, and a word processor that my father used to write at least onebook. There was only room in memory for about 2 pages of text, sohe'd write 2 pages at a time and then print them out, but it was alot better than a typewriter.Though I liked programming, I didn't plan to study it in college.In college I was going to study philosophy, which sounded much morepowerful. It seemed, to my naive high school self, to be the studyof the ultimate truths, compared to which the things studied inother fields would be mere domain knowledge. What I discovered whenI got to college was that the other fields took up so much of thespace of ideas that there wasn't much left for these supposedultimate truths. All that seemed left for philosophy were edge casesthat people in other fields felt could safely be ignored.I couldn't have put this into words when I was 18. All I knew atthe time was that I kept taking philosophy courses and they keptbeing boring. So I decided to switch to AI.AI was in the air in the mid 1980s, but there were two thingsespecially that made me want to work on it: a novel by Heinleincalled The Moon is a Harsh Mistress, which featured an intelligentcomputer called Mike, and a PBS documentary that showed TerryWinograd using SHRDLU. I haven't tried rereading The Moon is a HarshMistress, so I don't know how well it has aged, but when I read itI was drawn entirely into its world. It seemed only a matter oftime before we'd have Mike, and when I saw Winograd using SHRDLU,it seemed like that time would be a few years at most. All you hadto do was teach SHRDLU more words.There weren't any classes in AI at Cornell then, not even graduateclasses, so I started trying to teach myself. Which meant learningLisp, since in those days Lisp was regarded as the language of AI.The commonly used programming languages then were pretty primitive,and programmers' ideas correspondingly so. The default language atCornell was a Pascal-like language called PL/I, and the situationwas similar elsewhere. Learning Lisp expanded my concept of a programso fast that it was years before I started to have a sense of wherethe new limits were. This was more like it; this was what I hadexpected college to do. It wasn't happening in a class, like it wassupposed to, but that was ok. For the next couple years I was on aroll. I knew what I was going to do.For my undergraduate thesis, I reverse-engineered SHRDLU. My Goddid I love working on that program. It was a pleasing bit of code,but what made it even more exciting was my belief — hard to imaginenow, but not unique in 1985 — that it was already climbing thelower slopes of intelligence.I had gotten into a program at Cornell that didn't make you choosea major. You could take whatever classes you liked, and choosewhatever you liked to put on your degree. I of course chose "ArtificialIntelligence." When I got the actual physical diploma, I was dismayedto find that the quotes had been included, which made them read asscare-quotes. At the time this bothered me, but now it seems amusinglyaccurate, for reasons I was about to discover.I applied to 3 grad schools: MIT and Yale, which were renowned forAI at the time, and Harvard, which I'd visited because Rich Draveswent there, and was also home to Bill Woods, who'd invented thetype of parser I used in my SHRDLU clone. Only Harvard accepted me,so that was where I went.I don't remember the moment it happened, or if there even was aspecific moment, but during the first year of grad school I realizedthat AI, as practiced at the time, was a hoax. By which I mean thesort of AI in which a program that's told "the dog is sitting onthe chair" translates this into some formal representation and addsit to the list of things it knows.What these programs really showed was that there's a subset ofnatural language that's a formal language. But a very proper subset.It was clear that there was an unbridgeable gap between what theycould do and actually understanding natural language. It was not,in fact, simply a matter of teaching SHRDLU more words. That wholeway of doing AI, with explicit data structures representing concepts,was not going to work. Its brokenness did, as so often happens,generate a lot of opportunities to write papers about variousband-aids that could be applied to it, but it was never going toget us Mike.So I looked around to see what I could salvage from the wreckageof my plans, and there was Lisp. I knew from experience that Lispwas interesting for its own sake and not just for its associationwith AI, even though that was the main reason people cared aboutit at the time. So I decided to focus on Lisp. In fact, I decidedto write a book about Lisp hacking. It's scary to think how littleI knew about Lisp hacking when I started writing that book. Butthere's nothing like writing a book about something to help youlearn it. The book, On Lisp, wasn't published till 1993, but I wrotemuch of it in grad school.Computer Science is an uneasy alliance between two halves, theoryand systems. The theory people prove things, and the systems peoplebuild things. I wanted to build things. I had plenty of respect fortheory — indeed, a sneaking suspicion that it was the more admirableof the two halves — but building things seemed so much more exciting.The problem with systems work, though, was that it didn't last.Any program you wrote today, no matter how good, would be obsoletein a couple decades at best. People might mention your software infootnotes, but no one would actually use it. And indeed, it wouldseem very feeble work. Only people with a sense of the history ofthe field would even realize that, in its time, it had been good.There were some surplus Xerox Dandelions floating around the computerlab at one point. Anyone who wanted one to play around with couldhave one. I was briefly tempted, but they were so slow by presentstandards; what was the point? No one else wanted one either, sooff they went. That was what happened to systems work.I wanted not just to build things, but to build things that wouldlast.In this dissatisfied state I went in 1988 to visit Rich Draves atCMU, where he was in grad school. One day I went to visit theCarnegie Institute, where I'd spent a lot of time as a kid. Whilelooking at a painting there I realized something that might seemobvious, but was a big surprise to me. There, right on the wall,was something you could make that would last. Paintings didn'tbecome obsolete. Some of the best ones were hundreds of years old.And moreover this was something you could make a living doing. Notas easily as you could by writing software, of course, but I thoughtif you were really industrious and lived really cheaply, it had tobe possible to make enough to survive. And as an artist you couldbe truly independent. You wouldn't have a boss, or even need to getresearch funding.I had always liked looking at paintings. Could I make them? I hadno idea. I'd never imagined it was even possible. I knew intellectuallythat people made art — that it didn't just appear spontaneously— but it was as if the people who made it were a different species.They either lived long ago or were mysterious geniuses doing strangethings in profiles in Life magazine. The idea of actually beingable to make art, to put that verb before that noun, seemed almostmiraculous.That fall I started taking art classes at Harvard. Grad studentscould take classes in any department, and my advisor, Tom Cheatham,was very easy going. If he even knew about the strange classes Iwas taking, he never said anything.So now I was in a PhD program in computer science, yet planning tobe an artist, yet also genuinely in love with Lisp hacking andworking away at On Lisp. In other words, like many a grad student,I was working energetically on multiple projects that were not mythesis.I didn't see a way out of this situation. I didn't want to drop outof grad school, but how else was I going to get out? I rememberwhen my friend Robert Morris got kicked out of Cornell for writingthe internet worm of 1988, I was envious that he'd found such aspectacular way to get out of grad school.Then one day in April 1990 a crack appeared in the wall. I ran intoprofessor Cheatham and he asked if I was far enough along to graduatethat June. I didn't have a word of my dissertation written, but inwhat must have been the quickest bit of thinking in my life, Idecided to take a shot at writing one in the 5 weeks or so thatremained before the deadline, reusing parts of On Lisp where Icould, and I was able to respond, with no perceptible delay "Yes,I think so. I'll give you something to read in a few days."I picked applications of continuations as the topic. In retrospectI should have written about macros and embedded languages. There'sa whole world there that's barely been explored. But all I wantedwas to get out of grad school, and my rapidly written dissertationsufficed, just barely.Meanwhile I was applying to art schools. I applied to two: RISD inthe US, and the Accademia di Belli Arti in Florence, which, becauseit was the oldest art school, I imagined would be good. RISD acceptedme, and I never heard back from the Accademia, so off to ProvidenceI went.I'd applied for the BFA program at RISD, which meant in effect thatI had to go to college again. This was not as strange as it sounds,because I was only 25, and art schools are full of people of differentages. RISD counted me as a transfer sophomore and said I had to dothe foundation that summer. The foundation means the classes thateveryone has to take in fundamental subjects like drawing, color,and design.Toward the end of the summer I got a big surprise: a letter fromthe Accademia, which had been delayed because they'd sent it toCambridge England instead of Cambridge Massachusetts, inviting meto take the entrance exam in Florence that fall. This was now onlyweeks away. My nice landlady let me leave my stuff in her attic. Ihad some money saved from consulting work I'd done in grad school;there was probably enough to last a year if I lived cheaply. Nowall I had to do was learn Italian.Only stranieri (foreigners) had to take this entrance exam. Inretrospect it may well have been a way of excluding them, becausethere were so many stranieri attracted by the idea of studyingart in Florence that the Italian students would otherwise have beenoutnumbered. I was in decent shape at painting and drawing from theRISD foundation that summer, but I still don't know how I managedto pass the written exam. I remember that I answered the essayquestion by writing about Cezanne, and that I cranked up theintellectual level as high as I could to make the most of my limitedvocabulary. [2]I'm only up to age 25 and already there are such conspicuous patterns.Here I was, yet again about to attend some august institution inthe hopes of learning about some prestigious subject, and yet againabout to be disappointed. The students and faculty in the paintingdepartment at the Accademia were the nicest people you could imagine,but they had long since arrived at an arrangement whereby thestudents wouldn't require the faculty to teach anything, and inreturn the faculty wouldn't require the students to learn anything.And at the same time all involved would adhere outwardly to theconventions of a 19th century atelier. We actually had one of thoselittle stoves, fed with kindling, that you see in 19th centurystudio paintings, and a nude model sitting as close to it as possiblewithout getting burned. Except hardly anyone else painted her besidesme. The rest of the students spent their time chatting or occasionallytrying to imitate things they'd seen in American art magazines.Our model turned out to live just down the street from me. She madea living from a combination of modelling and making fakes for alocal antique dealer. She'd copy an obscure old painting out of abook, and then he'd take the copy and maltreat it to make it lookold. [3]While I was a student at the Accademia I started painting stilllives in my bedroom at night. These paintings were tiny, becausethe room was, and because I painted them on leftover scraps ofcanvas, which was all I could afford at the time. Painting stilllives is different from painting people, because the subject, asits name suggests, can't move. People can't sit for more than about15 minutes at a time, and when they do they don't sit very still.So the traditional m.o. for painting people is to know how to painta generic person, which you then modify to match the specific personyou're painting. Whereas a still life you can, if you want, copypixel by pixel from what you're seeing. You don't want to stopthere, of course, or you get merely photographic accuracy, and whatmakes a still life interesting is that it's been through a head.You want to emphasize the visual cues that tell you, for example,that the reason the color changes suddenly at a certain point isthat it's the edge of an object. By subtly emphasizing such thingsyou can make paintings that are more realistic than photographs notjust in some metaphorical sense, but in the strict information-theoreticsense. [4]I liked painting still lives because I was curious about what I wasseeing. In everyday life, we aren't consciously aware of much we'reseeing. Most visual perception is handled by low-level processesthat merely tell your brain "that's a water droplet" without tellingyou details like where the lightest and darkest points are, or"that's a bush" without telling you the shape and position of everyleaf. This is a feature of brains, not a bug. In everyday life itwould be distracting to notice every leaf on every bush. But whenyou have to paint something, you have to look more closely, andwhen you do there's a lot to see. You can still be noticing newthings after days of trying to paint something people usually takefor granted, just as you can afterdays of trying to write an essay about something people usuallytake for granted.This is not the only way to paint. I'm not 100% sure it's even agood way to paint. But it seemed a good enough bet to be worthtrying.Our teacher, professor Ulivi, was a nice guy. He could see I workedhard, and gave me a good grade, which he wrote down in a sort ofpassport each student had. But the Accademia wasn't teaching meanything except Italian, and my money was running out, so at theend of the first year I went back to the US.I wanted to go back to RISD, but I was now broke and RISD was veryexpensive, so I decided to get a job for a year and then return toRISD the next fall. I got one at a company called Interleaf, whichmade software for creating documents. You mean like Microsoft Word?Exactly. That was how I learned that low end software tends to eathigh end software. But Interleaf still had a few years to live yet.[5]Interleaf had done something pretty bold. Inspired by Emacs, they'dadded a scripting language, and even made the scripting language adialect of Lisp. Now they wanted a Lisp hacker to write things init. This was the closest thing I've had to a normal job, and Ihereby apologize to my boss and coworkers, because I was a bademployee. Their Lisp was the thinnest icing on a giant C cake, andsince I didn't know C and didn't want to learn it, I never understoodmost of the software. Plus I was terribly irresponsible. This wasback when a programming job meant showing up every day during certainworking hours. That seemed unnatural to me, and on this point therest of the world is coming around to my way of thinking, but atthe time it caused a lot of friction. Toward the end of the year Ispent much of my time surreptitiously working on On Lisp, which Ihad by this time gotten a contract to publish.The good part was that I got paid huge amounts of money, especiallyby art student standards. In Florence, after paying my part of therent, my budget for everything else had been $7 a day. Now I wasgetting paid more than 4 times that every hour, even when I wasjust sitting in a meeting. By living cheaply I not only managed tosave enough to go back to RISD, but also paid off my college loans.I learned some useful things at Interleaf, though they were mostlyabout what not to do. I learned that it's better for technologycompanies to be run by product people than sales people (thoughsales is a real skill and people who are good at it are really goodat it), that it leads to bugs when code is edited by too many people,that cheap office space is no bargain if it's depressing, thatplanned meetings are inferior to corridor conversations, that big,bureaucratic customers are a dangerous source of money, and thatthere's not much overlap between conventional office hours and theoptimal time for hacking, or conventional offices and the optimalplace for it.But the most important thing I learned, and which I used in bothViaweb and Y Combinator, is that the low end eats the high end:that it's good to be the "entry level" option, even though thatwill be less prestigious, because if you're not, someone else willbe, and will squash you against the ceiling. Which in turn meansthat prestige is a danger sign.When I left to go back to RISD the next fall, I arranged to dofreelance work for the group that did projects for customers, andthis was how I survived for the next several years. When I cameback to visit for a project later on, someone told me about a newthing called HTML, which was, as he described it, a derivative ofSGML. Markup language enthusiasts were an occupational hazard atInterleaf and I ignored him, but this HTML thing later became a bigpart of my life.In the fall of 1992 I moved back to Providence to continue at RISD.The foundation had merely been intro stuff, and the Accademia hadbeen a (very civilized) joke. Now I was going to see what real artschool was like. But alas it was more like the Accademia than not.Better organized, certainly, and a lot more expensive, but it wasnow becoming clear that art school did not bear the same relationshipto art that medical school bore to medicine. At least not thepainting department. The textile department, which my next doorneighbor belonged to, seemed to be pretty rigorous. No doubtillustration and architecture were too. But painting was post-rigorous.Painting students were supposed to express themselves, which to themore worldly ones meant to try to cook up some sort of distinctivesignature style.A signature style is the visual equivalent of what in show businessis known as a "schtick": something that immediately identifies thework as yours and no one else's. For example, when you see a paintingthat looks like a certain kind of cartoon, you know it's by RoyLichtenstein. So if you see a big painting of this type hanging inthe apartment of a hedge fund manager, you know he paid millionsof dollars for it. That's not always why artists have a signaturestyle, but it's usually why buyers pay a lot for such work.[6]There were plenty of earnest students too: kids who "could draw"in high school, and now had come to what was supposed to be thebest art school in the country, to learn to draw even better. Theytended to be confused and demoralized by what they found at RISD,but they kept going, because painting was what they did. I was notone of the kids who could draw in high school, but at RISD I wasdefinitely closer to their tribe than the tribe of signature styleseekers.I learned a lot in the color class I took at RISD, but otherwise Iwas basically teaching myself to paint, and I could do that forfree. So in 1993 I dropped out. I hung around Providence for a bit,and then my college friend Nancy Parmet did me a big favor. Arent-controlled apartment in a building her mother owned in NewYork was becoming vacant. Did I want it? It wasn't much more thanmy current place, and New York was supposed to be where the artistswere. So yes, I wanted it![7]Asterix comics begin by zooming in on a tiny corner of Roman Gaulthat turns out not to be controlled by the Romans. You can dosomething similar on a map of New York City: if you zoom in on theUpper East Side, there's a tiny corner that's not rich, or at leastwasn't in 1993. It's called Yorkville, and that was my new home.Now I was a New York artist — in the strictly technical sense ofmaking paintings and living in New York.I was nervous about money, because I could sense that Interleaf wason the way down. Freelance Lisp hacking work was very rare, and Ididn't want to have to program in another language, which in thosedays would have meant C++ if I was lucky. So with my unerring nosefor financial opportunity, I decided to write another book on Lisp.This would be a popular book, the sort of book that could be usedas a textbook. I imagined myself living frugally off the royaltiesand spending all my time painting. (The painting on the cover ofthis book, ANSI Common Lisp, is one that I painted around thistime.)The best thing about New York for me was the presence of Idelle andJulian Weber. Idelle Weber was a painter, one of the earlyphotorealists, and I'd taken her painting class at Harvard. I'venever known a teacher more beloved by her students. Large numbersof former students kept in touch with her, including me. After Imoved to New York I became her de facto studio assistant.She liked to paint on big, square canvases, 4 to 5 feet on a side.One day in late 1994 as I was stretching one of these monsters therewas something on the radio about a famous fund manager. He wasn'tthat much older than me, and was super rich. The thought suddenlyoccurred to me: why don't I become rich? Then I'll be able to workon whatever I want.Meanwhile I'd been hearing more and more about this new thing calledthe World Wide Web. Robert Morris showed it to me when I visitedhim in Cambridge, where he was now in grad school at Harvard. Itseemed to me that the web would be a big deal. I'd seen what graphicaluser interfaces had done for the popularity of microcomputers. Itseemed like the web would do the same for the internet.If I wanted to get rich, here was the next train leaving the station.I was right about that part. What I got wrong was the idea. I decidedwe should start a company to put art galleries online. I can'thonestly say, after reading so many Y Combinator applications, thatthis was the worst startup idea ever, but it was up there. Artgalleries didn't want to be online, and still don't, not the fancyones. That's not how they sell. I wrote some software to generateweb sites for galleries, and Robert wrote some to resize images andset up an http server to serve the pages. Then we tried to sign upgalleries. To call this a difficult sale would be an understatement.It was difficult to give away. A few galleries let us make sitesfor them for free, but none paid us.Then some online stores started to appear, and I realized thatexcept for the order buttons they were identical to the sites we'dbeen generating for galleries. This impressive-sounding thing calledan "internet storefront" was something we already knew how to build.So in the summer of 1995, after I submitted the camera-ready copyof ANSI Common Lisp to the publishers, we started trying to writesoftware to build online stores. At first this was going to benormal desktop software, which in those days meant Windows software.That was an alarming prospect, because neither of us knew how towrite Windows software or wanted to learn. We lived in the Unixworld. But we decided we'd at least try writing a prototype storebuilder on Unix. Robert wrote a shopping cart, and I wrote a newsite generator for stores — in Lisp, of course.We were working out of Robert's apartment in Cambridge. His roommatewas away for big chunks of time, during which I got to sleep in hisroom. For some reason there was no bed frame or sheets, just amattress on the floor. One morning as I was lying on this mattressI had an idea that made me sit up like a capital L. What if we ranthe software on the server, and let users control it by clickingon links? Then we'd never have to write anything to run on users'computers. We could generate the sites on the same server we'd servethem from. Users wouldn't need anything more than a browser.This kind of software, known as a web app, is common now, but atthe time it wasn't clear that it was even possible. To find out,we decided to try making a version of our store builder that youcould control through the browser. A couple days later, on August12, we had one that worked. The UI was horrible, but it proved youcould build a whole store through the browser, without any clientsoftware or typing anything into the command line on the server.Now we felt like we were really onto something. I had visions of awhole new generation of software working this way. You wouldn'tneed versions, or ports, or any of that crap. At Interleaf therehad been a whole group called Release Engineering that seemed tobe at least as big as the group that actually wrote the software.Now you could just update the software right on the server.We started a new company we called Viaweb, after the fact that oursoftware worked via the web, and we got $10,000 in seed fundingfrom Idelle's husband Julian. In return for that and doing theinitial legal work and giving us business advice, we gave him 10%of the company. Ten years later this deal became the model for YCombinator's. We knew founders needed something like this, becausewe'd needed it ourselves.At this stage I had a negative net worth, because the thousanddollars or so I had in the bank was more than counterbalanced bywhat I owed the government in taxes. (Had I diligently set asidethe proper proportion of the money I'd made consulting for Interleaf?No, I had not.) So although Robert had his graduate student stipend,I needed that seed funding to live on.We originally hoped to launch in September, but we got more ambitiousabout the software as we worked on it. Eventually we managed tobuild a WYSIWYG site builder, in the sense that as you were creatingpages, they looked exactly like the static ones that would begenerated later, except that instead of leading to static pages,the links all referred to closures stored in a hash table on theserver.It helped to have studied art, because the main goal of an onlinestore builder is to make users look legit, and the key to lookinglegit is high production values. If you get page layouts and fontsand colors right, you can make a guy running a store out of hisbedroom look more legit than a big company.(If you're curious why my site looks so old-fashioned, it's becauseit's still made with this software. It may look clunky today, butin 1996 it was the last word in slick.)In September, Robert rebelled. "We've been working on this for amonth," he said, "and it's still not done." This is funny inretrospect, because he would still be working on it almost 3 yearslater. But I decided it might be prudent to recruit more programmers,and I asked Robert who else in grad school with him was really good.He recommended Trevor Blackwell, which surprised me at first, becauseat that point I knew Trevor mainly for his plan to reduce everythingin his life to a stack of notecards, which he carried around withhim. But Rtm was right, as usual. Trevor turned out to be afrighteningly effective hacker.It was a lot of fun working with Robert and Trevor. They're the twomost independent-minded people I know, and in completely differentways. If you could see inside Rtm's brain it would look like acolonial New England church, and if you could see inside Trevor'sit would look like the worst excesses of Austrian Rococo.We opened for business, with 6 stores, in January 1996. It was justas well we waited a few months, because although we worried we werelate, we were actually almost fatally early. There was a lot oftalk in the press then about ecommerce, but not many people actuallywanted online stores.[8]There were three main parts to the software: the editor, whichpeople used to build sites and which I wrote, the shopping cart,which Robert wrote, and the manager, which kept track of orders andstatistics, and which Trevor wrote. In its time, the editor was oneof the best general-purpose site builders. I kept the code tightand didn't have to integrate with any other software except Robert'sand Trevor's, so it was quite fun to work on. If all I'd had to dowas work on this software, the next 3 years would have been theeasiest of my life. Unfortunately I had to do a lot more, all ofit stuff I was worse at than programming, and the next 3 years wereinstead the most stressful.There were a lot of startups making ecommerce software in the secondhalf of the 90s. We were determined to be the Microsoft Word, notthe Interleaf. Which meant being easy to use and inexpensive. Itwas lucky for us that we were poor, because that caused us to makeViaweb even more inexpensive than we realized. We charged $100 amonth for a small store and $300 a month for a big one. This lowprice was a big attraction, and a constant thorn in the sides ofcompetitors, but it wasn't because of some clever insight that weset the price low. We had no idea what businesses paid for things.$300 a month seemed like a lot of money to us.We did a lot of things right by accident like that. For example,we did what's now called "doing things that don't scale," althoughat the time we would have described it as "being so lame that we'redriven to the most desperate measures to get users." The most commonof which was building stores for them. This seemed particularlyhumiliating, since the whole raison d'etre of our software was thatpeople could use it to make their own stores. But anything to getusers.We learned a lot more about retail than we wanted to know. Forexample, that if you could only have a small image of a man's shirt(and all images were small then by present standards), it was betterto have a closeup of the collar than a picture of the whole shirt.The reason I remember learning this was that it meant I had torescan about 30 images of men's shirts. My first set of scans wereso beautiful too.Though this felt wrong, it was exactly the right thing to be doing.Building stores for users taught us about retail, and about how itfelt to use our software. I was initially both mystified and repelledby "business" and thought we needed a "business person" to be incharge of it, but once we started to get users, I was converted,in much the same way I was converted to fatherhood once I had kids.Whatever users wanted, I was all theirs. Maybe one day we'd haveso many users that I couldn't scan their images for them, but inthe meantime there was nothing more important to do.Another thing I didn't get at the time is that growth rate is theultimate test of a startup. Our growth rate was fine. We had about70 stores at the end of 1996 and about 500 at the end of 1997. Imistakenly thought the thing that mattered was the absolute numberof users. And that is the thing that matters in the sense thatthat's how much money you're making, and if you're not making enough,you might go out of business. But in the long term the growth ratetakes care of the absolute number. If we'd been a startup I wasadvising at Y Combinator, I would have said: Stop being so stressedout, because you're doing fine. You're growing 7x a year. Just don'thire too many more people and you'll soon be profitable, and thenyou'll control your own destiny.Alas I hired lots more people, partly because our investors wantedme to, and partly because that's what startups did during theInternet Bubble. A company with just a handful of employees wouldhave seemed amateurish. So we didn't reach breakeven until aboutwhen Yahoo bought us in the summer of 1998. Which in turn meant wewere at the mercy of investors for the entire life of the company.And since both we and our investors were noobs at startups, theresult was a mess even by startup standards.It was a huge relief when Yahoo bought us. In principle our Viawebstock was valuable. It was a share in a business that was profitableand growing rapidly. But it didn't feel very valuable to me; I hadno idea how to value a business, but I was all too keenly aware ofthe near-death experiences we seemed to have every few months. Norhad I changed my grad student lifestyle significantly since westarted. So when Yahoo bought us it felt like going from rags toriches. Since we were going to California, I bought a car, a yellow1998 VW GTI. I remember thinking that its leather seats alone wereby far the most luxurious thing I owned.The next year, from the summer of 1998 to the summer of 1999, musthave been the least productive of my life. I didn't realize it atthe time, but I was worn out from the effort and stress of runningViaweb. For a while after I got to California I tried to continuemy usual m.o. of programming till 3 in the morning, but fatiguecombined with Yahoo's prematurely agedculture and grim cube farmin Santa Clara gradually dragged me down. After a few months itfelt disconcertingly like working at Interleaf.Yahoo had given us a lot of options when they bought us. At thetime I thought Yahoo was so overvalued that they'd never be worthanything, but to my astonishment the stock went up 5x in the nextyear. I hung on till the first chunk of options vested, then in thesummer of 1999 I left. It had been so long since I'd painted anythingthat I'd half forgotten why I was doing this. My brain had beenentirely full of software and men's shirts for 4 years. But I haddone this to get rich so I could paint, I reminded myself, and nowI was rich, so I should go paint.When I said I was leaving, my boss at Yahoo had a long conversationwith me about my plans. I told him all about the kinds of picturesI wanted to paint. At the time I was touched that he took such aninterest in me. Now I realize it was because he thought I was lying.My options at that point were worth about $2 million a month. If Iwas leaving that kind of money on the table, it could only be togo and start some new startup, and if I did, I might take peoplewith me. This was the height of the Internet Bubble, and Yahoo wasground zero of it. My boss was at that moment a billionaire. Leavingthen to start a new startup must have seemed to him an insanely,and yet also plausibly, ambitious plan.But I really was quitting to paint, and I started immediately.There was no time to lose. I'd already burned 4 years getting rich.Now when I talk to founders who are leaving after selling theircompanies, my advice is always the same: take a vacation. That'swhat I should have done, just gone off somewhere and done nothingfor a month or two, but the idea never occurred to me.So I tried to paint, but I just didn't seem to have any energy orambition. Part of the problem was that I didn't know many peoplein California. I'd compounded this problem by buying a house up inthe Santa Cruz Mountains, with a beautiful view but miles fromanywhere. I stuck it out for a few more months, then in desperationI went back to New York, where unless you understand about rentcontrol you'll be surprised to hear I still had my apartment, sealedup like a tomb of my old life. Idelle was in New York at least, andthere were other people trying to paint there, even though I didn'tknow any of them.When I got back to New York I resumed my old life, except now I wasrich. It was as weird as it sounds. I resumed all my old patterns,except now there were doors where there hadn't been. Now when I wastired of walking, all I had to do was raise my hand, and (unlessit was raining) a taxi would stop to pick me up. Now when I walkedpast charming little restaurants I could go in and order lunch. Itwas exciting for a while. Painting started to go better. I experimentedwith a new kind of still life where I'd paint one painting in theold way, then photograph it and print it, blown up, on canvas, andthen use that as the underpainting for a second still life, paintedfrom the same objects (which hopefully hadn't rotted yet).Meanwhile I looked for an apartment to buy. Now I could actuallychoose what neighborhood to live in. Where, I asked myself andvarious real estate agents, is the Cambridge of New York? Aided byoccasional visits to actual Cambridge, I gradually realized therewasn't one. Huh.Around this time, in the spring of 2000, I had an idea. It was clearfrom our experience with Viaweb that web apps were the future. Whynot build a web app for making web apps? Why not let people editcode on our server through the browser, and then host the resultingapplications for them?[9]You could run all sorts of serviceson the servers that these applications could use just by making anAPI call: making and receiving phone calls, manipulating images,taking credit card payments, etc.I got so excited about this idea that I couldn't think about anythingelse. It seemed obvious that this was the future. I didn't particularlywant to start another company, but it was clear that this idea wouldhave to be embodied as one, so I decided to move to Cambridge andstart it. I hoped to lure Robert into working on it with me, butthere I ran into a hitch. Robert was now a postdoc at MIT, andthough he'd made a lot of money the last time I'd lured him intoworking on one of my schemes, it had also been a huge time sink.So while he agreed that it sounded like a plausible idea, he firmlyrefused to work on it.Hmph. Well, I'd do it myself then. I recruited Dan Giffin, who hadworked for Viaweb, and two undergrads who wanted summer jobs, andwe got to work trying to build what it's now clear is about twentycompanies and several open source projects worth of software. Thelanguage for defining applications would of course be a dialect ofLisp. But I wasn't so naive as to assume I could spring an overtLisp on a general audience; we'd hide the parentheses, like Dylandid.By then there was a name for the kind of company Viaweb was, an"application service provider," or ASP. This name didn't last longbefore it was replaced by "software as a service," but it was currentfor long enough that I named this new company after it: it was goingto be called Aspra.I started working on the application builder, Dan worked on networkinfrastructure, and the two undergrads worked on the first twoservices (images and phone calls). But about halfway through thesummer I realized I really didn't want to run a company — especiallynot a big one, which it was looking like this would have to be. I'donly started Viaweb because I needed the money. Now that I didn'tneed money anymore, why was I doing this? If this vision had to berealized as a company, then screw the vision. I'd build a subsetthat could be done as an open source project.Much to my surprise, the time I spent working on this stuff was notwasted after all. After we started Y Combinator, I would oftenencounter startups working on parts of this new architecture, andit was very useful to have spent so much time thinking about it andeven trying to write some of it.The subset I would build as an open source project was the new Lisp,whose parentheses I now wouldn't even have to hide. A lot of Lisphackers dream of building a new Lisp, partly because one of thedistinctive features of the language is that it has dialects, andpartly, I think, because we have in our minds a Platonic form ofLisp that all existing dialects fall short of. I certainly did. Soat the end of the summer Dan and I switched to working on this newdialect of Lisp, which I called Arc, in a house I bought in Cambridge.The following spring, lightning struck. I was invited to give atalk at a Lisp conference, so I gave one about how we'd used Lispat Viaweb. Afterward I put a postscript file of this talk online,on paulgraham.com, which I'd created years before using Viaweb buthad never used for anything. In one day it got 30,000 page views.What on earth had happened? The referring urls showed that someonehad posted it on Slashdot.[10]Wow, I thought, there's an audience. If I write something and putit on the web, anyone can read it. That may seem obvious now, butit was surprising then. In the print era there was a narrow channelto readers, guarded by fierce monsters known as editors. The onlyway to get an audience for anything you wrote was to get it publishedas a book, or in a newspaper or magazine. Now anyone could publishanything.This had been possible in principle since 1993, but not many peoplehad realized it yet. I had been intimately involved with buildingthe infrastructure of the web for most of that time, and a writeras well, and it had taken me 8 years to realize it. Even then ittook me several years to understand the implications. It meant therewould be a whole new generation of essays.[11]In the print era, the channel for publishing essays had beenvanishingly small. Except for a few officially anointed thinkerswho went to the right parties in New York, the only people allowedto publish essays were specialists writing about their specialties.There were so many essays that had never been written, because therehad been no way to publish them. Now they could be, and I was goingto write them.[12]I've worked on several different things, but to the extent therewas a turning point where I figured out what to work on, it waswhen I started publishing essays online. From then on I knew thatwhatever else I did, I'd always write essays too.I knew that online essays would be a marginal medium at first.Socially they'd seem more like rants posted by nutjobs on theirGeoCities sites than the genteel and beautifully typeset compositionspublished in The New Yorker. But by this point I knew enough tofind that encouraging instead of discouraging.One of the most conspicuous patterns I've noticed in my life is howwell it has worked, for me at least, to work on things that weren'tprestigious. Still life has always been the least prestigious formof painting. Viaweb and Y Combinator both seemed lame when we startedthem. I still get the glassy eye from strangers when they ask whatI'm writing, and I explain that it's an essay I'm going to publishon my web site. Even Lisp, though prestigious intellectually insomething like the way Latin is, also seems about as hip.It's not that unprestigious types of work are good per se. But whenyou find yourself drawn to some kind of work despite its currentlack of prestige, it's a sign both that there's something real tobe discovered there, and that you have the right kind of motives.Impure motives are a big danger for the ambitious. If anything isgoing to lead you astray, it will be the desire to impress people.So while working on things that aren't prestigious doesn't guaranteeyou're on the right track, it at least guarantees you're not on themost common type of wrong one.Over the next several years I wrote lots of essays about all kindsof different topics. O'Reilly reprinted a collection of them as abook, called Hackers & Painters after one of the essays in it. Ialso worked on spam filters, and did some more painting. I used tohave dinners for a group of friends every thursday night, whichtaught me how to cook for groups. And I bought another building inCambridge, a former candy factory (and later, twas said, pornstudio), to use as an office.One night in October 2003 there was a big party at my house. It wasa clever idea of my friend Maria Daniels, who was one of the thursdaydiners. Three separate hosts would all invite their friends to oneparty. So for every guest, two thirds of the other guests would bepeople they didn't know but would probably like. One of the guestswas someone I didn't know but would turn out to like a lot: a womancalled Jessica Livingston. A couple days later I asked her out.Jessica was in charge of marketing at a Boston investment bank.This bank thought it understood startups, but over the next year,as she met friends of mine from the startup world, she was surprisedhow different reality was. And how colorful their stories were. Soshe decided to compile a book of interviews with startup founders.When the bank had financial problems and she had to fire half herstaff, she started looking for a new job. In early 2005 she interviewedfor a marketing job at a Boston VC firm. It took them weeks to makeup their minds, and during this time I started telling her aboutall the things that needed to be fixed about venture capital. Theyshould make a larger number of smaller investments instead of ahandful of giant ones, they should be funding younger, more technicalfounders instead of MBAs, they should let the founders remain asCEO, and so on.One of my tricks for writing essays had always been to give talks.The prospect of having to stand up in front of a group of peopleand tell them something that won't waste their time is a greatspur to the imagination. When the Harvard Computer Society, theundergrad computer club, asked me to give a talk, I decided I wouldtell them how to start a startup. Maybe they'd be able to avoid theworst of the mistakes we'd made.So I gave this talk, in the course of which I told them that thebest sources of seed funding were successful startup founders,because then they'd be sources of advice too. Whereupon it seemedthey were all looking expectantly at me. Horrified at the prospectof having my inbox flooded by business plans (if I'd only known),I blurted out "But not me!" and went on with the talk. But afterwardit occurred to me that I should really stop procrastinating aboutangel investing. I'd been meaning to since Yahoo bought us, and nowit was 7 years later and I still hadn't done one angel investment.Meanwhile I had been scheming with Robert and Trevor about projectswe could work on together. I missed working with them, and it seemedlike there had to be something we could collaborate on.As Jessica and I were walking home from dinner on March 11, at thecorner of Garden and Walker streets, these three threads converged.Screw the VCs who were taking so long to make up their minds. We'dstart our own investment firm and actually implement the ideas we'dbeen talking about. I'd fund it, and Jessica could quit her job andwork for it, and we'd get Robert and Trevor as partners too.[13]Once again, ignorance worked in our favor. We had no idea how tobe angel investors, and in Boston in 2005 there were no Ron Conwaysto learn from. So we just made what seemed like the obvious choices,and some of the things we did turned out to be novel.There are multiple components to Y Combinator, and we didn't figurethem all out at once. The part we got first was to be an angel firm.In those days, those two words didn't go together. There were VCfirms, which were organized companies with people whose job it wasto make investments, but they only did big, million dollar investments.And there were angels, who did smaller investments, but these wereindividuals who were usually focused on other things and madeinvestments on the side. And neither of them helped founders enoughin the beginning. We knew how helpless founders were in some respects,because we remembered how helpless we'd been. For example, one thingJulian had done for us that seemed to us like magic was to get usset up as a company. We were fine writing fairly difficult software,but actually getting incorporated, with bylaws and stock and allthat stuff, how on earth did you do that? Our plan was not only tomake seed investments, but to do for startups everything Julian haddone for us.YC was not organized as a fund. It was cheap enough to run that wefunded it with our own money. That went right by 99% of readers,but professional investors are thinking "Wow, that means they gotall the returns." But once again, this was not due to any particularinsight on our part. We didn't know how VC firms were organized.It never occurred to us to try to raise a fund, and if it had, wewouldn't have known where to start.[14]The most distinctive thing about YC is the batch model: to fund abunch of startups all at once, twice a year, and then to spend threemonths focusing intensively on trying to help them. That part wediscovered by accident, not merely implicitly but explicitly dueto our ignorance about investing. We needed to get experience asinvestors. What better way, we thought, than to fund a whole bunchof startups at once? We knew undergrads got temporary jobs at techcompanies during the summer. Why not organize a summer program wherethey'd start startups instead? We wouldn't feel guilty for beingin a sense fake investors, because they would in a similar sensebe fake founders. So while we probably wouldn't make much money outof it, we'd at least get to practice being investors on them, andthey for their part would probably have a more interesting summerthan they would working at Microsoft.We'd use the building I owned in Cambridge as our headquarters.We'd all have dinner there once a week — on tuesdays, since I wasalready cooking for the thursday diners on thursdays — and afterdinner we'd bring in experts on startups to give talks.We knew undergrads were deciding then about summer jobs, so in amatter of days we cooked up something we called the Summer FoundersProgram, and I posted an announcement on my site, inviting undergradsto apply. I had never imagined that writing essays would be a wayto get "deal flow," as investors call it, but it turned out to bethe perfect source.[15]We got 225 applications for the SummerFounders Program, and we were surprised to find that a lot of themwere from people who'd already graduated, or were about to thatspring. Already this SFP thing was starting to feel more seriousthan we'd intended.We invited about 20 of the 225 groups to interview in person, andfrom those we picked 8 to fund. They were an impressive group. Thatfirst batch included reddit, Justin Kan and Emmett Shear, who wenton to found Twitch, Aaron Swartz, who had already helped write theRSS spec and would a few years later become a martyr for open access,and Sam Altman, who would later become the second president of YC.I don't think it was entirely luck that the first batch was so good.You had to be pretty bold to sign up for a weird thing like theSummer Founders Program instead of a summer job at a legit placelike Microsoft or Goldman Sachs.The deal for startups was based on a combination of the deal we didwith Julian ($10k for 10%) and what Robert said MIT grad studentsgot for the summer ($6k). We invested $6k per founder, which in thetypical two-founder case was $12k, in return for 6%. That had tobe fair, because it was twice as good as the deal we ourselves hadtaken. Plus that first summer, which was really hot, Jessica broughtthe founders free air conditioners.[16]Fairly quickly I realized that we had stumbled upon the way to scalestartup funding. Funding startups in batches was more convenientfor us, because it meant we could do things for a lot of startupsat once, but being part of a batch was better for the startups too.It solved one of the biggest problems faced by founders: theisolation. Now you not only had colleagues, but colleagues whounderstood the problems you were facing and could tell you how theywere solving them.As YC grew, we started to notice other advantages of scale. Thealumni became a tight community, dedicated to helping one another,and especially the current batch, whose shoes they remembered beingin. We also noticed that the startups were becoming one another'scustomers. We used to refer jokingly to the "YC GDP," but as YCgrows this becomes less and less of a joke. Now lots of startupsget their initial set of customers almost entirely from among theirbatchmates.I had not originally intended YC to be a full-time job. I was goingto do three things: hack, write essays, and work on YC. As YC grew,and I grew more excited about it, it started to take up a lot morethan a third of my attention. But for the first few years I wasstill able to work on other things.In the summer of 2006, Robert and I started working on a new versionof Arc. This one was reasonably fast, because it was compiled intoScheme. To test this new Arc, I wrote Hacker News in it. It wasoriginally meant to be a news aggregator for startup founders andwas called Startup News, but after a few months I got tired ofreading about nothing but startups. Plus it wasn't startup founderswe wanted to reach. It was future startup founders. So I changedthe name to Hacker News and the topic to whatever engaged one'sintellectual curiosity.HN was no doubt good for YC, but it was also by far the biggestsource of stress for me. If all I'd had to do was select and helpfounders, life would have been so easy. And that implies that HNwas a mistake. Surely the biggest source of stress in one's workshould at least be something close to the core of the work. WhereasI was like someone who was in pain while running a marathon notfrom the exertion of running, but because I had a blister from anill-fitting shoe. When I was dealing with some urgent problem duringYC, there was about a 60% chance it had to do with HN, and a 40%chance it had do with everything else combined.[17]As well as HN, I wrote all of YC's internal software in Arc. Butwhile I continued to work a good deal in Arc, I gradually stoppedworking on Arc, partly because I didn't have time to, and partlybecause it was a lot less attractive to mess around with the languagenow that we had all this infrastructure depending on it. So now mythree projects were reduced to two: writing essays and working onYC.YC was different from other kinds of work I've done. Instead ofdeciding for myself what to work on, the problems came to me. Every6 months there was a new batch of startups, and their problems,whatever they were, became our problems. It was very engaging work,because their problems were quite varied, and the good founderswere very effective. If you were trying to learn the most you couldabout startups in the shortest possible time, you couldn't havepicked a better way to do it.There were parts of the job I didn't like. Disputes between cofounders,figuring out when people were lying to us, fighting with people whomaltreated the startups, and so on. But I worked hard even at theparts I didn't like. I was haunted by something Kevin Hale oncesaid about companies: "No one works harder than the boss." He meantit both descriptively and prescriptively, and it was the secondpart that scared me. I wanted YC to be good, so if how hard I workedset the upper bound on how hard everyone else worked, I'd betterwork very hard.One day in 2010, when he was visiting California for interviews,Robert Morris did something astonishing: he offered me unsolicitedadvice. I can only remember him doing that once before. One day atViaweb, when I was bent over double from a kidney stone, he suggestedthat it would be a good idea for him to take me to the hospital.That was what it took for Rtm to offer unsolicited advice. So Iremember his exact words very clearly. "You know," he said, "youshould make sure Y Combinator isn't the last cool thing you do."At the time I didn't understand what he meant, but gradually itdawned on me that he was saying I should quit. This seemed strangeadvice, because YC was doing great. But if there was one thing rarerthan Rtm offering advice, it was Rtm being wrong. So this set methinking. It was true that on my current trajectory, YC would bethe last thing I did, because it was only taking up more of myattention. It had already eaten Arc, and was in the process ofeating essays too. Either YC was my life's work or I'd have to leaveeventually. And it wasn't, so I would.In the summer of 2012 my mother had a stroke, and the cause turnedout to be a blood clot caused by colon cancer. The stroke destroyedher balance, and she was put in a nursing home, but she reallywanted to get out of it and back to her house, and my sister and Iwere determined to help her do it. I used to fly up to Oregon tovisit her regularly, and I had a lot of time to think on thoseflights. On one of them I realized I was ready to hand YC over tosomeone else.I asked Jessica if she wanted to be president, but she didn't, sowe decided we'd try to recruit Sam Altman. We talked to Robert andTrevor and we agreed to make it a complete changing of the guard.Up till that point YC had been controlled by the original LLC wefour had started. But we wanted YC to last for a long time, and todo that it couldn't be controlled by the founders. So if Sam saidyes, we'd let him reorganize YC. Robert and I would retire, andJessica and Trevor would become ordinary partners.When we asked Sam if he wanted to be president of YC, initially hesaid no. He wanted to start a startup to make nuclear reactors.But I kept at it, and in October 2013 he finally agreed. We decidedhe'd take over starting with the winter 2014 batch. For the restof 2013 I left running YC more and more to Sam, partly so he couldlearn the job, and partly because I was focused on my mother, whosecancer had returned.She died on January 15, 2014. We knew this was coming, but it wasstill hard when it did.I kept working on YC till March, to help get that batch of startupsthrough Demo Day, then I checked out pretty completely. (I stilltalk to alumni and to new startups working on things I'm interestedin, but that only takes a few hours a week.)What should I do next? Rtm's advice hadn't included anything aboutthat. I wanted to do something completely different, so I decidedI'd paint. I wanted to see how good I could get if I really focusedon it. So the day after I stopped working on YC, I started painting.I was rusty and it took a while to get back into shape, but it wasat least completely engaging.[18]I spent most of the rest of 2014 painting. I'd never been able towork so uninterruptedly before, and I got to be better than I hadbeen. Not good enough, but better. Then in November, right in themiddle of a painting, I ran out of steam. Up till that point I'dalways been curious to see how the painting I was working on wouldturn out, but suddenly finishing this one seemed like a chore. SoI stopped working on it and cleaned my brushes and haven't paintedsince. So far anyway.I realize that sounds rather wimpy. But attention is a zero sumgame. If you can choose what to work on, and you choose a projectthat's not the best one (or at least a good one) for you, then it'sgetting in the way of another project that is. And at 50 there wassome opportunity cost to screwing around.I started writing essays again, and wrote a bunch of new ones overthe next few months. I even wrote a couple that weren't aboutstartups. Then in March 2015 I started working on Lisp again.The distinctive thing about Lisp is that its core is a languagedefined by writing an interpreter in itself. It wasn't originallyintended as a programming language in the ordinary sense. It wasmeant to be a formal model of computation, an alternative to theTuring machine. If you want to write an interpreter for a languagein itself, what's the minimum set of predefined operators you need?The Lisp that John McCarthy invented, or more accurately discovered,is an answer to that question.[19]McCarthy didn't realize this Lisp could even be used to programcomputers till his grad student Steve Russell suggested it. Russelltranslated McCarthy's interpreter into IBM 704 machine language,and from that point Lisp started also to be a programming languagein the ordinary sense. But its origins as a model of computationgave it a power and elegance that other languages couldn't match.It was this that attracted me in college, though I didn't understandwhy at the time.McCarthy's 1960 Lisp did nothing more than interpret Lisp expressions.It was missing a lot of things you'd want in a programming language.So these had to be added, and when they were, they weren't definedusing McCarthy's original axiomatic approach. That wouldn't havebeen feasible at the time. McCarthy tested his interpreter byhand-simulating the execution of programs. But it was already gettingclose to the limit of interpreters you could test that way — indeed,there was a bug in it that McCarthy had overlooked. To test a morecomplicated interpreter, you'd have had to run it, and computersthen weren't powerful enough.Now they are, though. Now you could continue using McCarthy'saxiomatic approach till you'd defined a complete programming language.And as long as every change you made to McCarthy's Lisp was adiscoveredness-preserving transformation, you could, in principle,end up with a complete language that had this quality. Harder todo than to talk about, of course, but if it was possible in principle,why not try? So I decided to take a shot at it. It took 4 years,from March 26, 2015 to October 12, 2019. It was fortunate that Ihad a precisely defined goal, or it would have been hard to keepat it for so long.I wrote this new Lisp, called Bel, in itself in Arc. That may soundlike a contradiction, but it's an indication of the sort of trickeryI had to engage in to make this work. By means of an egregiouscollection of hacks I managed to make something close enough to aninterpreter written in itself that could actually run. Not fast,but fast enough to test.I had to ban myself from writing essays during most of this time,or I'd never have finished. In late 2015 I spent 3 months writingessays, and when I went back to working on Bel I could barelyunderstand the code. Not so much because it was badly written asbecause the problem is so convoluted. When you're working on aninterpreter written in itself, it's hard to keep track of what'shappening at what level, and errors can be practically encryptedby the time you get them.So I said no more essays till Bel was done. But I told few peopleabout Bel while I was working on it. So for years it must haveseemed that I was doing nothing, when in fact I was working harderthan I'd ever worked on anything. Occasionally after wrestling forhours with some gruesome bug I'd check Twitter or HN and see someoneasking "Does Paul Graham still code?"Working on Bel was hard but satisfying. I worked on it so intensivelythat at any given time I had a decent chunk of the code in my headand could write more there. I remember taking the boys to thecoast on a sunny day in 2015 and figuring out how to deal with someproblem involving continuations while I watched them play in thetide pools. It felt like I was doing life right. I remember thatbecause I was slightly dismayed at how novel it felt. The good newsis that I had more moments like this over the next few years.In the summer of 2016 we moved to England. We wanted our kids tosee what it was like living in another country, and since I was aBritish citizen by birth, that seemed the obvious choice. We onlymeant to stay for a year, but we liked it so much that we stilllive there. So most of Bel was written in England.In the fall of 2019, Bel was finally finished. Like McCarthy'soriginal Lisp, it's a spec rather than an implementation, althoughlike McCarthy's Lisp it's a spec expressed as code.Now that I could write essays again, I wrote a bunch about topicsI'd had stacked up. I kept writing essays through 2020, but I alsostarted to think about other things I could work on. How should Ichoose what to do? Well, how had I chosen what to work on in thepast? I wrote an essay for myself to answer that question, and Iwas surprised how long and messy the answer turned out to be. Ifthis surprised me, who'd lived it, then I thought perhaps it wouldbe interesting to other people, and encouraging to those withsimilarly messy lives. So I wrote a more detailed version for othersto read, and this is the last sentence of it.Notes[1]My experience skipped a step in the evolution of computers:time-sharing machines with interactive OSes. I went straight frombatch processing to microcomputers, which made microcomputers seemall the more exciting.[2]Italian words for abstract concepts can nearly always bepredicted from their English cognates (except for occasional trapslike polluzione). It's the everyday words that differ. So if youstring together a lot of abstract concepts with a few simple verbs,you can make a little Italian go a long way.[3]I lived at Piazza San Felice 4, so my walk to the Accademiawent straight down the spine of old Florence: past the Pitti, acrossthe bridge, past Orsanmichele, between the Duomo and the Baptistery,and then up Via Ricasoli to Piazza San Marco. I saw Florence atstreet level in every possible condition, from empty dark winterevenings to sweltering summer days when the streets were packed withtourists.[4]You can of course paint people like still lives if you wantto, and they're willing. That sort of portrait is arguably the apexof still life painting, though the long sitting does tend to producepained expressions in the sitters.[5]Interleaf was one of many companies that had smart people andbuilt impressive technology, and yet got crushed by Moore's Law.In the 1990s the exponential growth in the power of commodity (i.e.Intel) processors rolled up high-end, special-purpose hardware andsoftware companies like a bulldozer.[6]The signature style seekers at RISD weren't specificallymercenary. In the art world, money and coolness are tightly coupled.Anything expensive comes to be seen as cool, and anything seen ascool will soon become equally expensive.[7]Technically the apartment wasn't rent-controlled butrent-stabilized, but this is a refinement only New Yorkers wouldknow or care about. The point is that it was really cheap, lessthan half market price.[8]Most software you can launch as soon as it's done. But whenthe software is an online store builder and you're hosting thestores, if you don't have any users yet, that fact will be painfullyobvious. So before we could launch publicly we had to launchprivately, in the sense of recruiting an initial set of users andmaking sure they had decent-looking stores.[9]We'd had a code editor in Viaweb for users to define theirown page styles. They didn't know it, but they were editing Lispexpressions underneath. But this wasn't an app editor, because thecode ran when the merchants' sites were generated, not when shoppersvisited them.[10]This was the first instance of what is now a familiar experience,and so was what happened next, when I read the comments and foundthey were full of angry people. How could I claim that Lisp wasbetter than other languages? Weren't they all Turing complete?People who see the responses to essays I write sometimes tell mehow sorry they feel for me, but I'm not exaggerating when I replythat it has always been like this, since the very beginning. Itcomes with the territory. An essay must tell readers things theydon't already know, and some people dislike being told such things.[11]People put plenty of stuff on the internet in the 90s ofcourse, but putting something online is not the same as publishingit online. Publishing online means you treat the online version asthe (or at least a) primary version.[12]There is a general lesson here that our experience with YCombinator also teaches: Customs continue to constrain you longafter the restrictions that caused them have disappeared. CustomaryVC practice had once, like the customs about publishing essays,been based on real constraints. Startups had once been much moreexpensive to start, and proportionally rare. Now they could be cheapand common, but the VCs' customs still reflected the old world,just as customs about writing essays still reflected the constraintsof the print era.Which in turn implies that people who are independent-minded (i.e.less influenced by custom) will have an advantage in fields affectedby rapid change (where customs are more likely to be obsolete).Here's an interesting point, though: you can't always predict whichfields will be affected by rapid change. Obviously software andventure capital will be, but who would have predicted that essaywriting would be?[13]Y Combinator was not the original name. At first we werecalled Cambridge Seed. But we didn't want a regional name, in casesomeone copied us in Silicon Valley, so we renamed ourselves afterone of the coolest tricks in the lambda calculus, the Y combinator.I picked orange as our color partly because it's the warmest, andpartly because no VC used it. In 2005 all the VCs used staid colorslike maroon, navy blue, and forest green, because they were tryingto appeal to LPs, not founders. The YC logo itself is an insidejoke: the Viaweb logo had been a white V on a red circle, so I madethe YC logo a white Y on an orange square.[14]YC did become a fund for a couple years starting in 2009,because it was getting so big I could no longer afford to fund itpersonally. But after Heroku got bought we had enough money to goback to being self-funded.[15]I've never liked the term "deal flow," because it impliesthat the number of new startups at any given time is fixed. Thisis not only false, but it's the purpose of YC to falsify it, bycausing startups to be founded that would not otherwise have existed.[16]She reports that they were all different shapes and sizes,because there was a run on air conditioners and she had to getwhatever she could, but that they were all heavier than she couldcarry now.[17]Another problem with HN was a bizarre edge case that occurswhen you both write essays and run a forum. When you run a forum,you're assumed to see if not every conversation, at least everyconversation involving you. And when you write essays, people posthighly imaginative misinterpretations of them on forums. Individuallythese two phenomena are tedious but bearable, but the combinationis disastrous. You actually have to respond to the misinterpretations,because the assumption that you're present in the conversation meansthat not responding to any sufficiently upvoted misinterpretationreads as a tacit admission that it's correct. But that in turnencourages more; anyone who wants to pick a fight with you sensesthat now is their chance.[18]The worst thing about leaving YC was not working with Jessicaanymore. We'd been working on YC almost the whole time we'd knowneach other, and we'd neither tried nor wanted to separate it fromour personal lives, so leaving was like pulling up a deeply rootedtree.[19]One way to get more precise about the concept of invented vsdiscovered is to talk about space aliens. Any sufficiently advancedalien civilization would certainly know about the Pythagoreantheorem, for example. I believe, though with less certainty, thatthey would also know about the Lisp in McCarthy's 1960 paper.But if so there's no reason to suppose that this is the limit ofthe language that might be known to them. Presumably aliens neednumbers and errors and I/O too. So it seems likely there exists atleast one path out of McCarthy's Lisp along which discoverednessis preserved.Thanks to Trevor Blackwell, John Collison, Patrick Collison, DanielGackle, Ralph Hazell, Jessica Livingston, Robert Morris, and HarjTaggar for reading drafts of this.
