 So, there was a time when I was putting my little pony or other cartoon characters in every talk I did, and it was a lot of fun. I don't have that this time, but I do have a story about a guy named Will. I was in grad school for a while, and my first year I lived by myself, which is great, particularly when you're out of college and you feel like you're an adult, but it was also very expensive. So, my second year I decided to get a roommate, and I went to the other people, I looked at the other people in the program, like, this guy, Will, he gets it. He'd be a great roommate. And he was nice, he was smart, he was, you know, friendly, he had just got out of a break-up so he needed a new roommate. And everything was great, we went home for the summer, I stayed on campus, I moved in the new apartment, I was waiting for Will, he got there, and then things started to emerge that maybe he wasn't such a great roommate that I thought, including the fact that this was one of his favorite books. Will was an objectivist, which may or may not mean things to you, but the talk both means he's an objectivist, it means that he was really selfish, he was not a great roommate. He was also progressively over this year becoming more and more pompous and grandiose, and he would make claims like, you know, I think I'm better than pretty much everybody, but more than average at pretty much anything, and we're like, Will, you don't speak Chinese, do you think you're better at speaking Chinese than average? Yeah. Yeah, I do. So in some ways, Will really predated the whole meritocracy culture of parts of San Francisco by thinking he was better than everybody. So we were at dinner one time with Will and my best friend Craig, who I then lived with after Will, because obvious reasons. And I don't remember what we were talking about, I desperately wish I could remember the subject of the conversation, but Will sort of checked out, and then when he checked back in, he checked back in with, you know, I can turn off my ability to read. And Craig and I just sort of looked at each other, because that makes no sense, right? Once you learn how to read, you can't not read. Like, if you know English, you can't not read this. It is psychologically impossible. And this quote stayed with me for obvious reasons over the years, and I never understood it until 15 years later. So I've been working as a developer for, I don't know, 10, 12 years at this point. I was sort of feeling burnt out on code, particularly as my only creative outlet. And so I went looking for what else to do. And I live in Durham, North Carolina, and I found the Durham Arts Council, which is a community center that has courses in everything from ballet to digital photography to improv comedy, which I did not take, to drawing. And I love drawing as a kid, so I thought, I'll try it. And I signed up for an introductory drawing course, and I really liked it. And I signed up for a figure drawing course, and I really liked it. And I signed up for a watercolor course, and I really liked it. And I kept going. So over the last almost three years now, I've been taking courses pretty regularly, going and learning how to become a better artist. And you can see, like, that was one of the introductory drawing courses. And some of these are okay. Some of them are not. I really like Sue the T-Rex from Chicago, if you guys know that. And one of the really neat things about these classes is that there was a huge diversity of people in them, drawn from around the triangle area, which is not huge, but it's fairly sizable. We had high schoolers, we had retirees, we had people of all different ethnic background, we had people who were professionals, we had people who worked manual labor and came straight from construction sites in some cases. And we had people who had a lot of experience drawing and people who weren't as experienced. And this last bit is particularly relevant, because I noticed that the mistakes I made when I was learning how to draw were the same mistakes that other people were making. And you see this in a lot of different places. I'm going to give you a few examples. One is faces. Very few people love to draw portraits. We find it difficult. And there are a lot of reasons why. But there are some particularly egregious mistakes that a lot of people make before they realize what's going on. The art I'm about to show you is from a website, our Tumblr, called Bad Celebrity Fanart. So prepare yourselves. So this is known as the short skull phenomenon. When we think of a face, we think of the features, right? We think of eyes, nose, mouth, sometimes ears if they're pointed, like spock or something. What we don't think about is that there's as much head above your eyes as there is below it. And so if you just go to draw a picture, you end up cutting off the top of the head, and then gently, delicately placing the hair back on top of the exposed brain, which makes for a deeply unsettling uncanny-valued photo. If you cover that up, this isn't that bad, it could obviously be better. But really, that is painful to look at. Here's another one. Yeah, so I'm pretty sure this is supposed to be David Tennant. This is the closest photo I could find of it. It might also be Conan O'Brien, possibly. We have a few issues here. Right, one, he's got football eyes. Nobody's eyes actually look like footballs with a target inside them. If you look over there, they're kind of close, but they're just different enough to get that uncanny-value of not being right. He's got monster teeth. Truth is outlined in a line that's heavy enough to imply some separation, which looks like he could take a really good bite out of your arm, and he has no bottom teeth, weird. And then the same thing happened with his hair, we call that spaghetti hair, where each strand was lovingly, artisanally hand-crafted. But nobody's hair looks like that. We all know that if I look at Chris, each one of his hairs is actually in each hair. He's not a Lego minifig with a plastic helmet of hair. As cool as that would be. Conan, where's your next costume Lego minifig hair? Just a suggestion. Anyway, what we see when we look at people is not each hair. We see shapes that have texture, and we understand that to mean hair. The problem with all these mistakes is that we let what we know to be the case. We know that each tooth is separate, and so we draw them separately, even if we can't see them separately. When we know that each hair is individual and distinct and deserving of its own unique hand-crafted touch, but when you do that, it doesn't look like hair anymore. We're letting what we know to be the case interfere with what we actually draw. Kids are a really interesting place to look for people learning how to draw. I have a nine-year-old and a five-year-old. My five-year-old, as of a couple of weeks ago, it's changed because five-year-olds changed instantly and on a dime. She had a very defined algorithm for drawing a dragon. This is her dragon. She starts by drawing the body. Then she draws the tail. Next come the wings, the mouth, the eyes, and last but not least, the eyelashes. A very important part of any dragon, as you all know, if you were a five-year-old and my daughter. The problem here is that her mental image of a dragon includes eyelashes at almost the same weight as wings. She draws them before she draws arms or legs or eating a cake or whatever else her dragon wants to do. It's cute, but it's not what we normally think of as a dragon. She has a unique mental representation of a dragon that she's trying to communicate, and it's different than most of us, and Roland can serve as an authority on that. This is a drawing not actually, this was done by me last night, but it's a representation of the sorts of stick figures that my kids drew when they were like three. Those kids go through this phase, right? So there's no body. The arms and legs come directly out of the head, which is, I think, a superior form of living. But then more interestingly, you may be subtle. You may have noticed that there are too many fingers. So kids' mental model of people is that they have fingers, and before they can count and even after they can count, they don't actually pay attention to how many fingers you have. I don't know that they're sure that everybody has the same number of fingers. I should have asked when they were young enough to answer me honestly instead of lying. So the way kids at this age draw hands is they draw a blob, and they start adding fingers, and when they either get tired or feel like it's about enough, they stop. And so you get sort of Cthulhu mythos, hug monsters waiting to encase you in their love. I think my personal record of counting fingers was my daughter, Morgan, when she was three or four, one hand had 27 fingers, and the other had 19. So yeah, that's a lot of fingers. But last I'd be seen as only picking on children. There's a psychology experiment where you're asked to draw a bike from memory. I don't know if any of you know this. It's really interesting and cool. An Italian digital artist sort of rediscovered this on his own, and he went out into the streets of Italy and asked people to draw bikes from memory, and he got some amazing results. This looks cool, right? It looks like a bike, but he's a digital artist. So he took all these sketches, and he went home, and he digitally rendered them. So you can see that this bike might have some problems. Yeah, so it's a little unbalanced. There's some issues. If you sit on it, I don't know how you're going to reach the pedals, and I don't know what the chain going to both tires is at all. But the tires themselves are really cool. I would love my bike to have those tires. So again, the idea is that we have mental representations of bikes. We know what bikes look like. We've seen thousands of bikes. But the parts that we think are important are what gets locked into our mental representation. We know that bikes have wheels. If you look at all the other sketches, you see that people, they almost always add the handlebars. Almost always notice I said that, because they don't always add the handlebars. They almost always add the seat. They often forget the pedals. So I guess you're like on a balance bike, just going down the street, it's awesome. Almost no one, including professional cyclists, draw the frame correctly, because it's not part of their representation of the bike. We have sort of a lossy representation of the world around us. We can only store certain things. And what we do or do not store is really important to pay attention to. So have any of you ever heard of urban sketching? No, awesome. Urban sketching is a worldwide movement where people go out into the urban environment, I guess, and they sketch. So it's very confusingly named, I suppose. The idea is that instead of like plein air painting where you're going out and painting a beautiful mountain, you're going into a human built environment and you're trying to capture that place, right? And one of the unique challenges with urban sketching over something like figure drawing or even drawing a still life is that you often have like lettering in words in the world. So this one is a piece I did in Durham. It's okay, right? I mean, it was, it's what it is. But the letters are legible. I was sort of offset a little bit, so I wasn't looking directly on, but the perspective wasn't crazy. When the perspective gets a lot harder, it's a lot harder to read the words, right? So up there in the corner, that's the National Museum of the American Indian, NDC. And it was really, really hard to get even that legible. Just because when we read something, we're not paying attention to like the shape of the letter forms unless you're a font designer. We're not paying, and in the real world when you go out into a city, almost no lettering is just like a line, right? So reading is lossy. This happens when we're out in the world reading signs. We don't pay attention to various aspects of it. This happens when we're reading text, right? Hopefully all of you have gotten at least halfway through. Hopefully some of you have found the error in this. Raise your hand if you found the error. Cool, yeah. So if you haven't, the, the, right? If you've ever been fooled by proofreading your own work, this is why. You know what it says. You don't have to read every word to know what it says, so you don't pay attention to every word. You lose information when you're encoding it into your head. This is really important, and if you're drawing individual letters, it shows up in really weird ways. So if you ever draw bubble letters or like letters with volume as opposed to the lines, most of them are fine. Like if I draw an A, do, do, do. It's the same shape as an A, and then I put a hole in the middle. If I draw an O, that one's really easy. You just draw a big O and then a little O inside it. S's are really hard. It's because the external contour doesn't match the shape of the letter that you're trying to draw. And it's because when we see an S, we know it's an S. I know what an S looks like. My brain mansplains to me, right? I know what you're seeing, but I know what it really is. So this can be got around when you're looking at languages in which you're not fluent. So presumably all of you can read this. Presumably fewer of you can read this, which says basically the same thing. If there are any Japanese speakers who want to correct it, catch me after, Joe Nen. Cool. Yeah, so this basically means the same thing. I know a little Japanese, and so I can like pick out kitsune, which means fox. And inu over there, which means dog. If I asked you to draw that first sentence to draw it instead of just write it, chances are you would just write it. You would probably also misspellings. There aren't none in that I didn't try to fool you twice. If I asked you to draw the second sentence, you'd probably do a much better job because you're not familiar. You don't imbue semantic meaning to the characters, right? They're just shapes to you if you don't know the language. And so you're going to do a much better job of replicating it. There's sort of a middle ground. This is the same sentence basically in Welsh. Are there any Welsh speakers? That's actually what I really want to know. No? That's... This totally says the same thing. I'm not making that up. But here, if I ask you to draw this, you're going to catch all the misspellings because you're going to... Since you don't know the words, you're going to be paying much more attention to the order of the letters, but you're still going to lose like the physical aspects of it. So if I put it in copper plate gothic, bold, you might miss off all the serifs because you're just going to be writing the letters as opposed to writing the words in the first place. But note, does anybody want to take a guess as to which word means dog in this last one? What? No. It's chi, right there. This means lazy, which I thought was kind of a fun thing to find out. So there is hope. When you're drawing, you are... You're in an internal conflict with your own brain trying to draw what you see instead of what's actually there, but there are ways to sort of get around it. And so somebody in the front row want to pull out their phone and start a timer. So we're going to do gesture drawings, which mean that I'm just going to be trying to capture the essence of Brit here. Thank you, our lovely model. And show you some techniques. Hello. Lots of stuff going on. This is a professional endeavor. I assure you. Cool. So you all can see, let's pick something fun like, where'd my charcoal go? There it is. Hey, look, it's the line. I totally just drew that. I don't know if you noticed. That was me. All right. Cool. So does somebody have a timer out? Can you yell every 20 seconds? No. Change. Just really loud and as imperiously as you can. All right. Let's go. All right. So one technique. Where's my pencil? Oh, that sucks. Maybe hold off on the change, the imperious change for just a minute. That's upside down. This is sweet. I wish my camera hadn't moved. There we go. Yeah. This isn't going to be awkward at all. Okay. Go back to superhero. So one technique is called negative space. So he's got a triangle in between his legs. I can draw that. And that sort of tells me how long his legs are in relation to the space in between them. This is super cool and awkward. So if I go up here, I can get the negative space between his arms and his body. And what that does is it tells me, go ahead and change poses. It tells me how long his arms are. Ooh, we got the thinker. Nice. The negative space again. Oh, fantastic. Joan and the superhero. So if we look for the negative spaces, this is the triangle in between his leg and his arm. Ooh, even better. Can you get that? Cool. All right. So it lets you quickly and more accurately than if I were just saying, I wonder how big his head is. Drawing the negative spaces. And make this one a minute. Why did I pick the running man? It lets you fix proportions, which are one of the most prominent places that people make mistakes in. So here, what I'm doing is actually not doing negative spaces, though. I'm sort of building up lines in an attempt to get volume and give a sort of appearance of three-dimensionality between sort of overlapping lines and darkening certain places to give shading. And it ends up looking not at all like Britt, but it still sort of looks like a person, right? Please say yes. That's my Jeb Bush moment. Feel free to clap or not? So here's something. You can actually unpause for a second. So here's something cool about charcoal. It's very movable, which means you can do things to shade that you can't not pose with a medium like ink. And we'll do that. No. Oh, yeah, you moved. It's fine, right? So ink is a very different medium than charcoal, which is very different from watercolor, which is very different from graphite. And part of the journey of learning to draw is knowing what you want to do, what you want to accomplish. I actually like the looking up. That's a good look. What you want to accomplish and then what you use to best accomplish it, right? So if I know I'm not going to have time to go back in, I might use ink more than charcoal. If I know I have a lot of time, then using charcoal or graphite or something that's movable becomes a lot more feasible. So here you can stop. Stop, thanks. So here I did. Here I've actually sort of constructed the head a little bit. So I can't see his entire head, but I know heads are the same shape almost all the time. And if you put construction lines, it lets you get around the illusion that you're seeing and draw what's actually there instead of what you think is there, right? So we're going to switch back. Oh, you can just drop that or not drop it. Lightly, gently, lovingly place it on the ground. Yeah, you're good. Thank you. All right. So these are really quick drawings. I only showed you maybe two techniques, maybe three, of how we sort of learned to get around that what we're seeing and what we think is there. There's hope in that if you do this a lot, you end up changing your internal representations of what the world looks like, and then it's easier to ignore it in favor of what you see. The point of this talk is a metaphor and not a literary metaphor. I'm actually going to say like and ask quite a bit. So feel free to yell at me afterwards for misusing the word metaphor, but I'm using it psychologically and so literally or in an English class. There are shallow and deep metaphors. Shallow metaphors are ones that sort of like spring out at you and you understand the similarities between the two things really quickly and then you also understand differences really quickly. Shallow metaphors often lead to arguments about where the boundaries lie. Like I don't think you can push the metaphor that far. I don't think you've pushed it that far enough. Negative spaces are really interesting shallow metaphor, I think for testing and TDD in that you're working outside the figure to constrain it in certain ways and to help you build the right thing when you get to drawing the figure. In testing, you're writing code outside the system and you're constraining the system in certain ways. They both break down in certain cases. If I'm in a superhero pose and I draw this negative space and then I draw this one but I don't get the relationship between them right, then my figure actually might be giant or might be tiny or might have one arm up here and one arm down here. Similarly, writing tests, if you were to write all the tests for your system before you wrote any of the application code, how correct do you think the application would be when you were done? The right thing to do is to go back and forth. You write tests, you write code, you write tests, you write code. You draw a negative space outside the figure, you work inside the figure. You go around, you draw across the body, you draw this negative space and you have a better relationship. You're better able to get the whole thing done. That's a shallow metaphor. There are others. I think you can make a metaphor between ink being immutable and immutable data. That's really fun. I'm sure there are more. But we're going to go on to deep metaphors, which are the more interesting ones. Deep metaphors are ones that might take a little longer to get but they're much more useful and they're generally broadly applicable beyond just the two things that you're making a metaphor between. And the deep metaphor for drawing is questioning yourself. Questioning what you think you see. Questioning what you think is there. It's like, I know there's another ear there but I can't see it but I know it's there so I'm just going to draw it anyway. That's going to end up not looking like the person unless they're the rare person who doesn't actually have two ears on the same side of their face. I guess Picasso might have been not happy with his talk. But taking away from drawing for just a minute, this is a classic example. You see a Unix system. I know Unix. I can know the system. How many of you have ever tried to use a co-worker's computer with their VIM setup with different plugins and dot files than you're used to? I think I can move on from this slide. So I worked at Living Social for a couple years. I ran their email systems. I maintained and built new ones and worked with a team and we sent more mail than almost anybody else at the time including many of your email service providers. I left Living Social and I went to a startup that had maybe 100,000 customers and soon after I was there we needed a new email system. Like, I know how to build an email system. The solution that worked for Living Social was not the solution that worked for Oneful. The solution that worked for either one of those was not the solution I went to when I went to another startup that was in the millions of customers and needed a new email solution. The problem is that we build up these middle models. We think we know how the world works. We think we know what the problem is just seeing the first tiny piece of it and so we extrapolate and we start to apply a solution and that's a mistake, right? Robert Wright is an evolutionary psychologist. He has this book out recently. It's a non-religious, non-spiritual version of Buddhism. He calls it humanistic Buddhism. And the idea is that... One of the main ideas apparently is that human suffering comes about because there's a mismatch between how we see the world and how it really is. That's the exact same problem we have in drawing. We're drawing what we think the world is instead of what it really is. It's the same problem we have in developing products where building products that we think people will use instead of the ones they'll actually use. Drawing is a really interesting and educational activity for me because it has these problems and it fails faster than almost anything else. I can have... Back up here. Don't worry about it. I hold up my picture next to him and I can see where I failed where what I drew was in my head but was not on him. If it's an hour drawing there might be more stuff that I did wrong or that I did right. We don't have that luxury so we have to be a lot more proactive in every other field. This is a book that came out about two weeks ago, Technically Wrong. It's a really, really good read. It's an exploration of how the preconceived notions of technologists and product people and startups in general result in things that are bad, that are actively bad. I highly recommend taking a look. But again, there's hope. With practice, with effort, we all can aspire to the greatness that Will had when he could turn off his ability to read. I still don't know if he was actually being honest. Probably not, knowing Will, but if we could get to that point where we can choose what layer to apply our mental models of the world onto what we're seeing, we'd be much more successful in everything. So in keeping with the season and weirdly on topic to this talk, any questions? Ask me during a break. Don't ask me now because that's not allowed. References and I mentioned it was weirdly apical to talk. These are two like oral histories of the David S. Pumpkin's phenomenon. Go take a look. Thank you.