 for coming by. I know it's been a long conference. Let's just make sure that we show the same level of energy participation enthusiasm on day four as well. My name is Tathagat Verma. I'm the track chair for Beyond Agile. I was not here last three days. I just got back to Bangalore today morning after a 48-hour flight. So I'm very happy to be here. It was a promise to Nareesh. I wanted to make sure I'll be here. So I'll just make a short welcome intro for Dave, our keynote speaker for the morning. Dave, as you all know, is one of the original signatories of Agile Manifesto back in 2001. He's also the author of Pragmatic Programmers. And the most remarkable thing that I think we want to make a big deal of is that Dave has not spoken at a conference in the last 14 years after signing up the Agile Manifesto. So Dave, it's our honor to host you here today morning. And we look forward to the keynote. Can we give a big round of applause to Dave here? Please welcome Dave. Thank you for that. I appreciate it. Yeah, actually, I want to talk about that, too, because first of all, what a fantastic conference. I'm really enjoying this. Great speakers, a fantastic location. And mostly really interesting conversations. I mean, most conferences, a lot of the real cool stuff takes place, you know, outside the sessions. And I found it's absolutely fascinating just talking to everybody here over the last three, four days. So thank you very much for having me. Really appreciate it. Yeah, so I was one of the people who helped create the Manifesto, what was it, 13 years ago, 14 years ago. And since then, I have deliberately not been involved with any Agile events. So I've not been to any of the Agile conferences and haven't participated in the Agile Alliance. And there's a reason for it. I think that agility is absolutely necessary, not just in software, but in everything we do. But I think once you give something a name, it's almost inevitable that people take it over. I mean, you name something, it becomes real estate. And then people can start like making a claim to parts of it. So today I want to do two things. I have two talks, which is good for you, right? Two for the price of one. The bad thing is the first one is one of my rents, right? I'm angry and I'm going to get this out and then we'll get over it and we'll have a nice pleasant talk, all right? So with your permission, we're going to have a rant. For which I shall probably be damned, but whatever. Two parts to this, two parts. The first one is I am so, so, so, so sick of people talking about Agile. All right? Can we stop doing this, please? There is no such thing as Agile. Agile is not a noun. You don't have a lump of Agile sitting somewhere, just waiting to be used. But people use it all the time, all right? We get book proposals sent to us. People want to write books called things like Pragmatic Agile. What? All right? So at its best, Agile is an adjective, okay? So you can use Agile to say that, you know, this is an Agile process. That's kind of okay, all right? I'll let you do that. You have my permission to do that. But even better, please, just stop using the word Agile because the key point here is that Agility, which is an adverb, better describes it because Agility is how you do things, right? So stop with the nouns and start with the verbs. Stop saying this is an Agile something. Instead say, I just did this with Agility because that better describes the original spirit of the movement. Let me show you something which has Agility. I hope this video actually works. Here we go. This is a two-wheeled, self-balancing little robot. It is wandering over a fairly rough gravel surface in a circle. It's balancing itself. It finds obstacles and gets around them. Inside, it is using a pretty straightforward, very standard control system algorithm, which basically is a feedback loop. It basically says to itself, hey, am I falling over? Yeah, I better move my wheels a bit faster to get where I want to be. Now, it's more complicated than that because it also has to take into account the fact it wants to move forward, and it has to take into account some of the history of what's taking place so it knows how effective the various movements are. But basically, that is Agility. That's what I want to see everybody here doing. Maybe, maybe not sure what we should do is go get a whole bunch of unicycles and train everybody to ride around on unicycles so they can learn to be agile that way. But then, what happens is everybody says, okay, but how do I use this? How do I use this in my job? I want to take something back, give me a bullet list of things I can do when I get back. Right? So that's really hard. So what do we do? We bring in the consultants. All right? Consultants are people who basically are going to take money to do the things that you would have done anyway, but they give you the courage to do them. So you would ask a consultant, for example, hey, I like this agile thing, but it's very small. It doesn't scale. So consultants or maybe vendors are going to go away and think about this for a bit and say, oh, we've got the answer to a two-wheeled little robot that doesn't scale. Here, we'll give you one of these. Is that agile? No, but it can stump on people if you want it to. This is where I totally screw up next year's conference. I apologize. But why are the tool vendors and the consultants here? First of all, they're not all evil, but why are they here? They're here because you're afraid. You're here because in your head, you're kind of worried. In your heart, you know what you want to do, but in your head, you're worried about this, this agility, agile, agility, whatever it is thing. All right? How do I do this? How do I do this? I don't know. You do know, but you don't believe it. So you have to bring people in to show you how to do it. Please have courage. In XP, one of the first agile processes, one of the, well, it's a varying number of process practices is courage. And that's always seemed like a strange thing to put in to a software development process, but actually it's one of the most important things because courage is what lets you do the right thing. And that's what I want you to do. Have courage. Stay true to the basics. That slide is all you need to know about agility. Everything else is just marketing. Where are we? Where do we want to be? Let's make a small move. Then we go back up the top and do it again. And we nest these loops. We do these loops on, what are we going to call this variable? And we do these loops on, how are we going to develop this product? There's lots and lots of loops all nested inside each other. But that's all there is to it. So let's kind of step back a little bit from the industrialization of the agile manifesto. And let's make it back into the human process that it is. So I want you to have courage and I also want you to have fun because if you're not having fun, you're not doing it right. So I'm sorry there's not a single bullet point in there of things you can take back, but I honestly think that's one of the most important takeaways you can come away from this conference with. So end of the rant. Sorry about that. I just had to get it off my chest. Now we can have the nice relaxing fun part. Let's all just, or I need to breathe just for a second. My talk starts with a prologue like any good story would. My prologue starts in the year 2002. Now I'm English, but I live in the States. And obviously the U.S. was rocked pretty hard by the events of September 2011, 2001. And in 2002 things were getting a little bit scary. We were in the middle of the quote, war on terror. Now the war on terror, you know how you name things which is kind of like the opposite of what they are just to make yourself feel brave? Well the war on terror was actually a whole group of people running around being really scared. And it was kind of like a really strange time to be. Fortunately, in the midst of all this we had Donald Rumsfeld, Secretary of Defense. A man, look at that, a man that you wouldn't want to mess with. Interesting, you know, talking about naming things the opposite of what they are, Secretary of Defense, for someone that goes out and basically chooses who you're going to bomb is a bit of a strange title as well. Anyway, Secretary Rumsfeld was widely quoted for something he said in 2002. He said, talking about information gathering and terrorists, there are known knowns. There are things we know that we know. There are unknown, sorry known unknowns. That is to say there are things that we now know we don't know. There are also unknown unknowns. There are things we do not know we don't know. And to a lot of people listening to that, it sounded like the word know about 100,000 times. It sounded silly, so they all kind of went, ho, ho, Donald Rumsfeld, what an idiot, well actually there's nothing wrong with this. It's a perfectly reasonable statement. Except as computer people, we know there's something missing from this statement. Because he talks about known knowns, known unknowns, and unknown unknowns. So what's missing? Right. So that's what I want to talk about. I want to talk about unknown knowns. This is my way of sneaking another manifesto on the world. Yay! Another opportunity for consultancy. Great. All right. So I want to talk about the tacit manifesto. And I'm going to do it by talking about four thinkers, four people who I think have some really profound things to say about the way we think about things, the way we do things. It used to be you go to conferences and give talks and you'd put up pictures of people and you'd say, who's this? Right. And you'd put up pictures of people like Alan Kay or Don Nooth or someone like this. And people would start guessing those. So after a while, you'd start putting up more and more and more obscure pictures, just because you didn't like the audience to beat you. Unless you can actually read the subtitle there. No one's going to know this guy. This guy is Bill Phillips. Bill Phillips was born in New Zealand. His father was an engineer of some sort. And he spent his weekends in the garage tinkering with various engineering projects. He is famous as an economist. But he actually had a whole range of careers before becoming an economist. My favorite being crocodile hunter. Does he look like a crocodile? I don't know. So the story is that he was born in New Zealand. He went to Australia, did his crocodile hunting, became a cinema manager and then qualified as an electrical engineer. War broke out. He went to fight. He was interned. He spent his time in a prisoner of war camp doing things like building secret radios using his engineering knowledge to do this. After the war finished, he decided that he had seen a whole bunch of the bad side of people. And he wanted to find out why. So he went to London to become a sociologist. And he went to the London School of Economics as a student. He could not stop tinkering though. He liked the engineering part. So in his spare time, he built a machine. And the machine was based on the fact we all know that money flows like water. Well, his genius was to say, hey, what? I could make water flow like money. And he invented this machine. It was called the Moniac as in money, something, something computer. Because at that time, you know, all the computers we used were named somethingiac. So he named it Moniac. But in reality, what he did is he invented an entire new discipline of academics called hydraulic macroeconomics. Is that cool or what? What it actually is is a system of pumps and tubes and tanks and valves that models the UK economy. It might be easier to understand as a diagram. Not really. So what it actually does is you have an amount of water in there that is circulating around, and the valves will sort of divert it depending on things like what's the current interest rates. So you can like say, more or less depends on how you divert the water. You can inject new money into it to represent inflation. You can tell it about, you know, balance of trade. You can tell it about unemployment, all these kind of things by just adjusting valves. And the water just pumps around and around and around just the way it does with real money in a real economy. And he demonstrated this machine to a colloquium at the London School of Economics. And based on that, he got a job on the staff and he became an economist. This machine became quite famous. There are, I think, like six left in the world. The one that works is back in New Zealand in the foyer of one of the banks. There's one in the Science Museum in London. I don't know where the other ones are. But one of the cool things about it, let me just show you on this picture, you can see at the top there, there's like a piece of paper with lines on them. Well, that's actually genuinely a piece of paper that's motorized. It moves from left to right. And there's a pen that's attached to a float in the tank. So he can graph the amount of water, the level of water in that tank over time. And doing that, he could graph whatever economic factor that tank happened to represent. So this machine would sit there, pump around and around and around, and graph out your economy. And here's the wild thing. At the time, it was the most accurate predictor of the economy that existed. In general, it was accurate to 2%. Or so they say. So you could feed this thing parameters, and it would predict what your economy would look like. No one else could do that at the time. No one else could solve in anywhere close to real time the equations that were involved in doing this. But the fun thing about this is no one could say why. It produced this output, right? It had nice graphs that looked pretty. And it was accurate. But no one could actually say, yes, this is why it's accurate. Apart from looking at it at something like saying, oh, yeah, you've got all the various factors involved. But no one could actually prove it was correct. But it was. Also interestingly, you could start with exactly the same initial conditions, and it would produce different output, sometimes slightly different output, sometimes very different output. So it would give you unexpected answers. And one of the challenges of using this machine is, is this a miracle answer, right? This unexpected answer I got. Is this some deep insight, or is it just wrong? And you wouldn't know. You had no way of knowing. Except you'd have to trust the fact that in general, it was pretty accurate. So I think that's a really, really cool machine. They use these things for many years until finally, electronic computers could easily replace them. I think that the Moniac behaves just like an expert. You feed it information, comes out with an answer, you don't know how it does it, and mostly it's right. But when it comes out with something kind of strange, you're never too sure. Is this right? Is this not right? Okay. Bill Phillips. We'll get back to him in a minute. I said we're going to be talking about fourth thinkers. This time we're going to get through two in one section. We're going to look at Stuart and Hubert Dreyfuss. Looks very much like, you know, their hairdressers here, but they're not. Stuart and Hubert Dreyfuss were, actually they were psychologists, but also early computer programmers. And one of the things they wanted to do was to model, is to teach machines to learn. But they realized they didn't actually know how people learned. So how are you going to teach a machine how to learn until you understand how people learn? So they got involved in this idea of digging into the mechanism by which you and I gather information. And this was discovered by the United States Air Force who wanted to do research into how to best train pilots. And so they commissioned the Dreyfuss brothers to produce a report. And for the longest time, I couldn't find it, but actually a couple of months back I actually found this report online. It's been declassified. I'll put the slides up. You'll see the URL here. It's an interesting read. It really is. So this is the report. It's titled A Five-Stage Model of the Mental Activities Involved in Directed Skills Acquisition. All right. How do we gain a skill? Their premise was that everybody goes through five stages on their journey to expert. And these five stages, they gave them names. The names aren't that important, but novice, competent, proficient, expert, and master. Now, these are not kind of like absolute stages. You don't come home from work one day and say, hey, guess what, honey? I'm proficient. It doesn't work that way. It's a continuum. But there are these five overall levels. And they looked at certain attributes that kind of help identify what level you're in. And they came up with four different axes to look at. Recollection, recognition, decision, and awareness. These aren't the best names. What they really mean by this is under the recognition, under the recollection column, what it really means is, do you use the context when you're solving a problem? All right. So this is actually pretty straightforward. If you're a novice, you've never seen any of this stuff before. You don't have any context. So all of your decisions are going to be context free. All of your decisions are going to be like looking just at what you see right now. But once you've gained even a little bit of experience, then the chances are you'll have seen some of this stuff before. So you can start applying context. The next thing is whether or not you're in a position to pattern match on that context. Can you recognize certain characteristics of what's going on? Have you seen something like this before or maybe two things that combine to form a third thing? Can you start synthesizing your own solutions based on the stuff that you're seeing going on? The next column is where it starts getting just a little bit controversial. And this is where the Air Force kind of bounced this out for a while before deciding, yes, it's actually correct. Under the decision column, the question they're answering is, are you working analytically? Hands up, everybody who thinks it's a good thing to work analytically. The rest of you thought it must be a trap. Their decision was it's not a good thing to work analytically. When you work analytically, you sit there and you try and think of all of the different factors. You try and think of all the different things you've seen. You try and apply all of this information. And then you somehow have to come up with some kind of weighted matrix that says what to do in these circumstances given this and that and the other. And then you come up with the one best solution. The problem with this, well, there's two problems. One is there often isn't a best solution. But the main problem is that takes a lot of time. You have to think about all of those things. And time is the one thing that you don't have if you're flying an airplane and something starts going wrong. So the Dreyfus Brothers said that at some point, you need to switch away from acting analytically and start trusting your instincts. And then finally, they have this awareness column. And that's really simple. The awareness column says, are you consciously monitoring yourself? Or are you unconsciously monitoring yourself? Because you're always looking at what you're doing. The question is, are you actually having to do that consciously? Are you going to have to stop what you do and look at what you're doing? Or is that happening in the background? So they produce this big matrix of all the various words, but you can actually simplify it down to this. Or you can even push it further. There you go. We come down to this. Novice has nothing. They're novices. Who cares? They just don't exist really. However, once you become competent, then your knowledge is situational in that you actually have context you can apply to solve the problem. They say that when you're proficient, you actually start to look at the whole big picture when you're solving these problems. If you're an expert, you have stopped thinking analytically. And you have started thinking intuitively. And finally, when you're a master, you have become totally absorbed in what you're doing. You don't interrupt yourself to look at yourself. It's all part of what you do. Nowadays, we call that flow. But this is the dry first model of skills acquisition. And it's useful, I think, in a number of ways. Firstly, if you're teaching somebody, I think it's really vital to understand the stages they're going to go through. If you're teaching an absolute beginner, for example, they have no context. They have no basis on which to make decisions apart from common sense. So it's actually very, very valid to tell them what to do. Because that way you're helping them get context for the next time. Beginners need to be told what to do. If I am going to jump out of an airplane, I don't really care about Bernoulli effect and the theory of aerodynamics and air pressure gradients and all this kind of stuff. All I want to know is what do I pull on my parachute and when do I pull it? I want someone to tell me what to do. That's what novices are. As you get more and more advanced, you start moving away from that. You're not telling them what to do anymore, but you're kind of entering into a partnership with them. It also is interesting to look at this in terms of how do I deal with people. Have you ever spoken to someone who's an expert or maybe worked with someone who's an expert? And you've written some code or you've come up with some idea and they say, I don't think that's a good idea. And you say, why not? And they say, I don't know. It just doesn't feel like a good idea. That is an expert talking because they really honestly don't know. They've had years of experience building patterns in their head. So many patterns. There's no way they can remember them all. But this particular circumstance you're in with that expert triggers some pattern. And they go, yeah, alarm bells are going off. I don't know why, but alarm bells are going off. Now, if you are an expert and you're also in industry or a consultant, then there's one skill that you have to have. And that is, you need to learn how to lie because you can't get away with saying to some manager somewhere, I don't know. So instead what you do is you say, oh, my instinct says this is bad. And then they say, why? And then you come up with some really plausible sounding reason. There's absolutely nothing to do with the actual reality which is buried somewhere in your head and you can't get it out. Now, interestingly, I want to talk about that expert phase. Alfred North Whitehead said civilization advances by extending the number of important operations we can perform without thinking. I think that is absolutely crucial. This is what is called tacit knowledge. It's knowing something without knowing that you know it. It's when the knowledge is buried deep enough that it just comes forward. And we all have this, we all have this in so many different ways. From the moment we're born, we are developing tacit knowledge. Have you ever watched a baby when they're very, very little, move their arms and their legs? It looks random. They're just like flailing about all the time. Well, what they're actually doing is they're slowly building a picture in their brain of movement, of how thinking certain things affects how their hands and legs move. And gradually over time, you'll see them starting to practice not just random stuff but actually touching individual things. As their knowledge grows, they no longer have to control all those individual movements. And instead, they can start putting together more macro movements like reach for the bear or whatever it might be. They get that ability because they are building tacit knowledge. So the fourth person I want to talk about is this gentleman, Michael Poliani. He was, well, look at this. I mean, he wasn't a crocodile hunter, but he had a pretty good resume. Physical, chemist, sociologist, economist, and philosopher. He's probably best known as a philosopher. He wrote a very, well, actually what he did is he gave a series of three very influential talks that were put into a book called The Tacit Dimension. And the book is about how far science can go. And it's about how science has to rely on tacit knowledge. It's about the unknowns. His point really is how do you know that something is even possible if you don't know about it? Because that's what scientists have to do every day. They have to imagine something that doesn't exist yet. They have to form a theory about it and then try to work out how to get there based on various theories. So they're constantly dealing with this unknown, with this tacit information. They have a gut feeling that things should work this way and then they have to work towards it. The first essay in this book I think is the one that's most significant for us. And in it he says, we know more than we can tell. Inside all of us is an incredible amount of knowledge that we cannot put a voice to. We cannot describe. And he has a theory about how this comes about. In his theory, we are all surrounded the entire time, day and night, by lots of different stimuli. In here is the sound of my voice. There's the light. There's my hands waving around. There's the air temperature. There's all sorts of things that are constantly going on that are stimulus. And most of these are kind of like below your level of consciousness. For example, the presence of gravity is probably not something you thought about this morning. You're all sitting there quite comfortable in the fact that you're sitting. Gravity has not entered the picture for you yet. But how many people have been in an earthquake? Yeah? How quickly do you suddenly remember gravity when the chair just comes out from underneath you? Just a little bit. Suddenly that's like the most important thing in the world to you. That stimulus is always being recorded. But normally it's just in the background. You're not thinking about it. It's not part of your consciousness. But then what happens is something that you are aware of comes in. Something that is actually conscious. It's a sensed effect. And the two of these things are happening at the same time. And your brain, stupid organ that it is, says, ah, there's a causality here. In this set of circumstances, in this set of stimuli, then this particular effect can happen. And so you form this causal link. And he calls this causal link a tacit knowledge of the stimuli. That is, you're not consciously aware of integrating all of these things together. But your brain has made that connection. They have done some remarkably ugly experiments to prove this. One is they had people sit down with electrodes on them. And they read these people just a continuous series of words. One or two of those words were associated with electric shock. So they'd just be reading the words, you know, apple, banana, pear, car, whatever. And when they hit one of the words that was associated with shock, they would push a button and the subject would receive a small or hopefully small electric shock. And they did this for however long it took to condition the people. And then they would say to people when they finished the experiment, they would say, okay, so what did you happen? And the people would say, well, I heard you doing this stream of words. And every now and then I got a shock. And they'd say, well, why did you get the shock? And the people would say, I don't know. What caused the shock? I don't know. So then part two of the experiment is they would use the same electrodes, but rather than using them to shock people, they would use them to sense muscle activity. They would sit them down and they would read the same stream of words to them. When they got to the words that triggered the shock, people tensed. They had no idea they were doing it. They didn't realize those words caused the shock. But in their head, the connection was made. So they were building this tacit knowledge inside their head without thinking about it. Other things that we do tacitly, one of the classic examples, is how do you recognize people? When I got here and I was sitting out at the front there and Nourish walked by. All I had seen of him was a picture about that big on a web page. And in my head, I thought, that's probably Nourish. It was like 60, 40, 70, 30, that kind of thing. That's probably him. So I just tried it and said, hey, Nourish. And he was like, oh, yeah. All right. Ask me how I did that. I have no idea. How do you do it? How do you recognize faces? Or if you see somebody on the street walking away from you, can you recognize them by their walk? Yes, you can. How? We don't know. But you've got the experience built into tacit knowledge that lets you do that. We could not exist if we didn't have all of that tacit knowledge rumbling around inside our head. So you can think of tacit knowledge as being an integration of our experiences, the things that we do without thinking. Think how bad life would be if we couldn't use this. I learned to fly airplanes when I came to the States because I'd always wanted to do it, and it was really expensive in England. It was cheaper in the States. So I thought, OK, this is my opportunity. Of course, clever me. I got my license just before 9-11, and then they changed all the rules. But while I was learning, I was absolutely amazed. You get into playing with a pilot and you fly around. On your very, very first lesson, the chances are you will land that airplane. Obviously, your instructor has got his hands very, very close to the controls, but you will land the airplane. And typically, after maybe six hours of instruction, at some point, the pilot will fly you out to an airfield somewhere in the middle of nowhere. You'll land the plane. He'll tell you to pull over, stop, and then he will get out, or she will get out, and say, OK, now it's your turn. So you solo after a very small number of hours. Five, six, seven, whatever it might be. And yet, you don't get your license for another maybe 40 or 50 hours. And why is that? It's because on that very first solo, your brain is the busiest it has ever been in its life. Your brain is sitting there thinking about every single movement that your hands make, that your feet make. It's thinking about every single instrument in front of you. It's looking out the car. It is just like absolute total overload. All you can do is follow the instructions that you've learned to take this thing off, fly in a circle, and land it. That's all your brain can do. So for the next 40 hours, what you're learning to do is to move that from your consciousness into your subconscious, your tacit layer. So you can start flying that airplane without thinking about it, at least flying all the mechanical stuff, until at the point where you can maintain straight and level flight without even thinking about it. You're just automatically moving your hands. And that's not magical. Everybody who drives does that. Everybody who drives a car can keep the car going without thinking about it. You can leave your house and go somewhere, and when you get to that other place, you may not even remember driving. It's so buried down deep in your brain. If you've ever taught someone to drive, you know exactly how that is. When you're teaching someone to drive, the last thing they're going to do is have a conversation with you. Because they're just like, and you're sitting there, oh, it's nice today, isn't it? Knowledge has to be tacit to be usable. The cool thing about tacit knowledge is that it's cumulative. When you start off as an infant, you don't have very much of it because you don't have any experience or much experience anyway. So I don't know how learning to speak and understand speech happens, but it could well be something like this. Initially, every sound you hear, you're hearing. It's conscious. Everything, speech, the noise in the room, cars, whatever it might be, are just sounds. And you don't really have any kind of processing for them. It's just like trying to, oh, something strange. But eventually, sounds start to get processed automatically. And once they do, then you can start processing words. Again, initially, you have to think about them. After a while, it becomes tacit. Then the same thing happens to phrases because we don't listen to each other speaking word by word by word. We listen to whole sentences and parse them in our heads. The interesting thing is, if you're learning a language, you actually go through those phases. My wife's learning Chinese at the moment. She's learning Mandarin. And I kind of like tagged along a little bit until my brain gave out. But I was very conscious of the fact that initially, all I was hearing was sounds. And I wasn't hearing them very well. And I started to say to myself, is that a sound? I was consciously having to do that, which meant I couldn't understand anything anyone said because by the time I'd worked that out, they were at the end of the sentence. They were well past me. But eventually, my brain started to hear the differences. And I didn't have to think about them anymore. And then I kind of got to the point where I could understand certain words. And that's kind of okay. You can actually maybe think of a sentence and say, oh, they talked about this, this, this and this. But I still didn't understand because I hadn't reached the point where I could parse a string of words into a phrase. My wife's reached that point. And now she's working on the next level of tacit knowledge, which is idiom, where you're not just worried about the words and their literal meaning, but she's worried about the meaning in context. All of this is tacit knowledge. All of this is knowledge that we build. One more example. Inner tennis. This was a thing in the 1970s. And it was a response. I think the guy's name was Galloway. He was a tennis coach. And he worked out that the reason that many people fail at tennis is because they get in their own way. They think too much about what they're doing. So he had a training system which sounds crazy. He would take people out onto a tennis court and say he was teaching them how to serve. He would take them onto a tennis court. On the other side of the net, he would place a chair, a regular old chair. And the person he was teaching would be on the opposite side of the court. He'd give them a big old bucket of tennis balls. And he would say, I want you to hit the ball over the net. I do not want you to try to hit the chair. I just want you to hit the ball over the net. But every time the ball lands, a ball lands on the other side of the net, I want you to say, is this ball to the left of the chair, to the right of the chair, in front of the chair, or behind the chair? So every time you hit the ball, I want to hear two words, left behind, right in front. And people would hit 100 balls, 200 balls, however many it took, and just go, left behind, right behind, left behind. And do that over and over again. At some point, he would then say, okay, hit the chair. And every single ball the person hit from then on would hit the chair. Their brains had somehow, in the same way the baby flailing around works out that that means that this hand goes left, their brains had accumulated this tacit knowledge of the effect of hitting the ball where the ball hit on the other side of the net. And once they'd trained their brain that way, then they could just choose to hit the chair every time. Tacit knowledge is the only way we can use knowledge effectively. Because tacit knowledge is how we operate on auto pilots, how we give ourselves room to think about what we're doing. Once you start thinking about the lower levels, you can't do the higher levels. And you can prove this to yourself. As an experiment when you leave here, probably not in the crowd, as you walk, don't just walk. Instead, command your legs to walk. Right? Tell your legs how to walk. You know how to move your muscles, right? I can move this muscle and that one and move my leg like that. I can control my leg by moving my muscles. Yeah? Everyone can. You can all do this. So control your muscles to make yourself walk. Try it. Or if you're talking to someone, as you're talking, focus on making the right sounds. I guarantee you'll stop talking. But when you're programming, stop using all of these layers of abstraction that you have. And instead, I want you to think about precisely what the computer is going to do down at the register level with every single thing that you type. And see just how effectively you apply the solid principles and all these other great design things in that environment. Just try it and see. The point I'm making is that the only viable knowledge that we have is tacit knowledge. And yet, we're all here trying to gain knowledge by trying to gain experience from other people. We're all at this conference listening to experts as they talk to us about this and the other. And we go away and we're always going to be slightly disappointed that we go away and it all seems to evaporate. I heard this really great talk about lean start-ups. Oh, really? What was it about? Well, I think there was something about minimum something or others. And it's never quite as good after you've left the conference. Why is that? It's because what you've done is you've listened to somebody else's tacit knowledge. You've listened to somebody else's experience. So to make use of this conference, to make use of any piece of learning, I think we have to recognize that there's billions of thinkers, not just four. We are thinkers. That's actually you. And we are all thinkers. But more than that, we are actually machines that will turn thinking into action. And the only way to do that is through experience. So if you want to learn from the stuff that you actually hear at a conference like this, you have to go and apply it. You have to go and exercise your brain. You have to go and gather this tacit knowledge through experience. So to be honest, this is somewhat ironic, but I just thought I'd put it in there anyway. I am a strong believer in tacit knowledge. I'm a strong believer in just trusting yourself. I believe that having experience is more important than knowing facts. I believe that any tool that enhances feedback is more important than a tool that enhances productivity. Maybe we need editors that give you an electric shock whenever you make a typing mistake. I believe that incremental, participative learning, wow, is more important than formal tuition. One way of expressing that is apprenticeships beat university degrees when it comes to the software. But there's many other areas. People, for example, learn on the job. They learn through incremental participation. If they're a novice, you start them off gently. You don't throw them in the deep end. They're working alongside people. They're gaining experience by doing things with other people. And gradually they're kind of developing on their own until, like a butterfly, they can fly away and help other people. And most important, I believe in intuition over rules. There are so many studies that say that people's intuition beats out any set of rules you can put in place because the rules are based on one flat scenario. But people with experience, experts who can apply intuition, can synthesize their own rules given any environment. That's a really, really hard sell in most companies. But the companies that succeed are the companies that give people space to use their gut, to use their intuition, to try things. Using your intuition means making mistakes. So the companies that succeed don't punish those mistakes. But if you're going to do something really different, you have to rely on intuition. So, upshot of all this, it says, teach yourself to learn. Actually, what it really means is give yourself permission to learn. That sounds very California New Age. What it really means is just try stuff. Try stuff over and over again. Try different things. Don't make a religion out of it. Don't try and judge how effective the trying is. Just do it. Trust just like with the people hitting the balls over the net. Just do it. Because your brain in the background is learning from what you're doing all the time. You don't have to sit there unconsciously say, what did I learn? Just do it and let your brain come out with tacit knowledge. And then, once you have learned, once you have got experience, I think each and every one of us has a sacred duty. And that is, we need to teach others. Because otherwise, all that experience dies with us. And the race as a whole doesn't get any better. So, use your experience to teach. But here is the trick. Don't teach people what you were taught. Instead, teach them what you have learned. Because that way, everybody benefits. Infuse them with what you've learned and inspire them with what you've learned. And that way, we'll create a new generation that will astound us in future. Thank you. Great talk. Curious about tacit knowledge at the team and organizational level. Do you have much to say on that? Sorry, that was tacit knowledge at a team? Yeah, of the team and the organizational level. Most of the examples have been at the individual. Yeah, actually, that's very true. And there's actually good and bad there. Every team has tacit knowledge. And typically, that tacit knowledge is embedded in the phrase, that's not how we do things around here. And that's not necessarily a good thing. There are, however, teams where the tacit knowledge is shared and shared accurately. The best example that comes to my mind right now is airline pilots, where they're trained, and it takes 12, 15,000 hours, but they spend so long together that they can actually work, almost like a married couple, picking up each other's mistakes, working together tacitly. They'll be making the call-outs and everything else, but below the surface, they'll be actually communicating. There's a really great example of that. Let's see if I can remember the details. I think it was a US Airways flight, I may be wrong, flying along 35,000 feet, immense explosion in the back. It turns out that the engine and tail exploded. And it threw out all of the rotor blades from the turbines. And those rotor blades did something which was, in theory, impossible. And that is, they took out all of the hydraulics in the aircraft. Now, the aircraft have three independent hydraulic systems. So in theory, you're always supposed to have hydraulics, because they control the flaps and everything else, right? Without them, you can't fly. Well, it turns out that all three hydraulic systems have to go through a fairly narrow channel to get into the tail. So when the tail engine blew, it severed all three. So the pilots were left on the plane that they could not control, right? The pedals didn't work, nothing worked. And so every pilot is trained in an emergency, not to panic, but instead, you go through some steps, you go through checklists. And one of the things you do is you turn to the red section in the pilot operating handbook for the aircraft to say, what do I do now? And the red section for total hydraulic failure was basically pray and say goodbye to people. Because the pilot of this plane, a guy called Al Haynes, I think he had like 25,000 hours flying. And he was lucky he had with him another senior pilot, because normally they fly inexperienced. And this time they had him and an experienced pilot, and they had a training pilot deadheading, sitting in the cockpit with him. So they had three, between them like 60,000 hours worth of flying experience. And they refused to die. So they tried to work out how they could fly this plane. And there are a number of secondary effects that the throttles have on an aircraft. So for example, if you throttle up, then initially a plane is going to climb. If you throttle down, it's going to descend. If you differentiate throttle, it's going to turn a little bit. But as it turns, it's also going to go up or down, depending on what's so new. Nothing has ever been written in any rule book about how to fly a plane using just the engines. These three pilots between them, and without saying very much, experimented with the controls until all three of them, the guy in the deadheading guy, was managing the throttles. Al Haynes, I can't remember what he was doing, and the copilot was handling some of the other systems. And the three of them between them actually managed to fly this aircraft, descend from 35,000 feet, take it down, and actually land at Sioux City, Iowa. Now the problem they had was to land, you have to turn the engines off effectively, so you lose control. So they came in, they lined up perfectly with the runway, they touched down, but one of the wings was a little bit low, so they're playing cartwheel. And if you ever get a chance, go look at the video, it's the most heart-stopping video you'll ever see. This plane touches down at cartwheels, breaks into, I think, three pieces, and all you see is this ball of flame, and you think, I've just watched 100 people die. But in fact, the tacit knowledge and the tacit cooperation between those three pilots saved the vast majority of people on that plane. They walked away from it, and it was considered impossible. So it is possible to be a tacit team, but I think to do it requires exactly the same commitment that it takes to get tacit knowledge as an individual. And that is you have to work together, you have to work for a long time, and you have to trust each other. And I think that's pretty rare in the industry, unfortunately. I don't think there's any shortcuts. Tacit knowledge gets passed on from in a social environment such as a team by being together and experiencing things together, as you just described. But doesn't it happen faster if you convert that tacit knowledge to explicit knowledge, such as what is happening here in the conference? And I am able to express and explicitly transfer what I know to you. And in that case, my experience would probably get communicated to you faster than if you were just to stay with me and observe me and learn from what I'm doing. I think there's an element of being able to learn tacit knowledge, but I don't think it's as significant as it might be. In the States, I think that some of the best-selling magazines and the best-selling books are golf books. God help us. And golf books are all about how to improve your swing, how to improve your short game. And they're full of these incredibly detailed things about, well, now you need to move your left foot just about a quarter inch forward and increase your stance like this now. It's like they're trying to tell you how to move. In the same way, if I try to tell you how to walk, we just put one foot in front of the other. No, you don't. There's a lot more to it than that. So I think that real, honest tacit knowledge can only be gained through experience. Now, what can happen is that you can guide that experience in the same way that parents guide the experience that their children have. And they try to maximize the learning that they can get out of different experiences. Then I think, yes, you can do the same thing in a context like this. So for example, if someone talks about a technique or a tool or whatever it might be at this conference, then simply knowing that that exists and then going away and trying it for yourself and whatever else, that will help you get tacit knowledge more quickly because it guides you where you want to go. But just telling you about it is not going to give you the knowledge that you need. All that's going to do is give you a goal. It's up to you to get that tacit knowledge. And I don't think there's a shortcut. I can accelerate it simply by guiding. And like I said, if you're working, if you're a novice, working alongside an expert or someone, and doesn't even an expert, someone slightly higher than you, it's called situated learning. If you can do that, you will gain experience more quickly. And that's why, for example, pair programming often works really, really well. Because pair programming, both sides are learning tacitly as they're going along. But I don't think you can just speak knowledge. And there's a word for people that speak knowledge. And that's consultant. It just doesn't always work. Oh, it's my fault. I'm a computer person. I'm sorry.