Startupfest 2012 - Coefficients of friction

Editor's Notes

• #2: I’m going to talk about coefficients of friction\n
• #3: Or, more specifically, why lube is better than thrust, and why the future is lube.\n
• #4: I’m writing a book with Ben Yoskovitz, announced last week, that will touch on many of the things I discuss in this presentation. We’d love it if you’d visit this URL, sign up and join the conversation.\n
• #5: Okay, so here we go.\nYour brain is greedy. Even though it’s only 2.5% of your weight, it consumes 20% of your food and oxygen.\n
• #6: When you’re born, it’s small—and grows really fast. This is because the size of our heads barely fits through our mothers’ pelvises.\n
• #7: We were, quite literally, born to think.\n
• #8: We have so many things we want to do\n
• #9: We want to build soaring monuments\n
• #10: Plumb the depths of the abyss\n
• #11: Or fly through the air\n
• #12: But the real world gets in the way of those plans\n
• #13: As it turns out, while ideas might be limitless, the world around us is full of messy atoms, and they’re a pain to deal with. They have friction.\n
• #14: We’re just not that good at most things.\n
• #15: Our spirit is eager, but our flesh is weak.\n
• #16: We have great ideas, and lousy containers with which to make them manifest.\n
• #17: You’re simply not that good at dealing with the real world and its limitations.\n
• #18: So we’ve set about building things, from the earliest tools.\n
• #19: Much of our enterprise has been about moving rocks around\n
• #20: Or learning to breathe underwater for a while\n
• #21: Or figuring out how to fly through the sky.\n
• #22: We didn’t have a hive mind, but we invented printing and broadcast\n
• #23: And voting to reach consensus (at least, that’s how it’s supposed to work.)\n
• #24: The lever, the pulley, the wheel were all ways of overcoming the limitations of our containers\n
• #25: A horse might be stronger than a human, but given enough block and tackle, a human wins every time.\n
• #26: Most of human history has been about overcoming the limitations of our containers. And most of that has been about fighting friction with thrust.\n
• #27: And that’s why the industrial era was such a transformation of humanity.\n
• #28: The invention of things like the steam engine were going to change everything in our fight against friction.\n
• #29: Steam engines gave us muscle without the complaints.\n
• #30: This forever changed how we dealt with the atoms that were between us and our ideas.\n
• #31: It led to an explosion of innovation. We went from an agrarian, feudal society where land was the key, to one where the means of production—and eventually mass production—were key.\n
• #32: Organizations got big, and competed on their ability to do things at scale, because when you’re fighting friction, more thrust is better.\n
• #33: Size has its advantages, to be sure.\n
• #34: Big companies compete based on this scale\n
• #35: It means you can control the means of production or raw materials\n
• #36: Train the workforce to doing things your way\n
• #37: Restrict or dictate how capital flows to the market\n
• #38: Get exclusivity for a process or brand\n
• #39: Control how the goods reach customers\n
• #40: Or change the regulatory and legislative environment\n
• #41: When it comes to moving atoms around, these are great ways to keep out competitors.\n
• #42: When it comes to moving atoms around, these are great ways to keep out competitors.\n
• #43: When it comes to moving atoms around, these are great ways to keep out competitors.\n
• #44: But big means slow to adapt, which is why you see phone companies, banks, carmakers and airlines asking for government bailouts and dictating lousy terms to their customers and partners.\n
• #45: It’s also why corporate org charts are so restrictive: they’re about span of control. Daniel McCallum is the father of management theory. He fixed railroads by bringing in military structure to organizations. “large armies, vertical integration and functional divisions under centralized command”\n
• #46: But something interesting happened on the way to the steam age.\n
• #47: In 1712, a British engineer named Newcomen developed a steam-powered pump that would pull water from coalmines. It quite literally fueled the steam age.\n
• #48: Between 1763 and 1775, James Watt improved on the design significantly.\n
• #49: Watt’s design was much more efficient—it 75% less coal to generate the same energy.\n
• #50: This made economist William Stanley Jevons’ wonder what was going on.\n
• #51: Shouldn’t consumption go down as efficiency increases? The reverse was happening.\n
• #52: This happens everywhere. A server today does the work of two a couple of years ago, and four a couple of years before that, because of Moore’s Law.\n
• #53: This happens everywhere. A server today does the work of two a couple of years ago, and four a couple of years before that, because of Moore’s Law.\n
• #54: This happens everywhere. A server today does the work of two a couple of years ago, and four a couple of years before that, because of Moore’s Law.\n
• #55: That means today’s server is a ROW of servers ten years ago\n
• #56: That means today’s server is a ROW of servers ten years ago\n
• #57: That means today’s server is a ROW of servers ten years ago\n
• #58: And a DATA CENTER back at the start of the Web.\n
• #59: And a DATA CENTER back at the start of the Web.\n
• #60: And a DATA CENTER back at the start of the Web.\n
• #61: Think about that for a minute. Today’s server is yesterday’s DATA CENTER.\n
• #62: Jevons would like to know, then: why are we still building data centers?\n
• #63: Jevons’ paradox, as it’s called, is that an increase in efficiency tends to increase demand.\n
• #64: Gates envisioned a computer in every household\n
• #65: But his vision had unintended consequences—a computer in every pocket, which was also a phone, a camera, a game console, a typewriter, a GPS, a personal physician, and more.\n
• #66: Now let’s talk about that explosion of software, and the abundance it creates.\n
• #67: Mark Andreesen’s a pretty smart guy. Recently, he said “software eats everything.” Which I take to mean that whenever there’s an input and an output that are digital, everything in the middle becomes software.\n
• #68: Mark Andreesen’s a pretty smart guy. Recently, he said “software eats everything.” Which I take to mean that whenever there’s an input and an output that are digital, everything in the middle becomes software.\n
• #69: Mark Andreesen’s a pretty smart guy. Recently, he said “software eats everything.” Which I take to mean that whenever there’s an input and an output that are digital, everything in the middle becomes software.\n
• #70: Mark Andreesen’s a pretty smart guy. Recently, he said “software eats everything.” Which I take to mean that whenever there’s an input and an output that are digital, everything in the middle becomes software.\n
• #71: I like to restate this as “lube beats thrust.”\n
• #72: Business models based on overcoming friction, are crumbling everywhere. Like the travel agency\n
• #73: Or the record store\n
• #74: Or the video store\n
• #75: Competition today isn’t about more thrust. It’s about eliminating the friction in the first place.\n
• #76: Companies that continue to think their value is more thrust—rather than less friction—are doomed.\n
• #77: The industrial age gave us the ability to move atoms, so our ideas could become real. But the digital age gives us the ability to ignore the atoms, and work purely with ideas.\n
• #78: Abundance is a big word. Here’s what I mean.\n
• #79: This is Doc Searls, one of the authors of the Cluetrain Manifesto. He talks about the economics of abundance, and how we can make money because something is abundant.\n
• #80: Sure, scarcity is valuable—that’s why gold and diamonds cost a lot. But abundance has value, too.\n
• #81: Think about Linux. IBM went from selling operating systems, to making money because of free ones.\n
• #82: Monty Python found that by making their videos free on Youtube, sales increased more than tenfold. They sold DVDs because their stuff was available for free.\n
• #83: Or bloggers. Plenty of people (myself included) don’t make money from their blog, but do make money because of it.\n
• #84: (With a quick shout-out to my co-author, Ben Yoskovitz. He credits his entire career to blogging; he even has an infographic to prove it. Ben, Jevon, and the GoInstant team sold their 14-person company to Salesforce 2 years after founding it for $76M. Incredible.)\n
• #85: (With a quick shout-out to my co-author, Ben Yoskovitz. He credits his entire career to blogging; he even has an infographic to prove it. Ben, Jevon, and the GoInstant team sold their 14-person company to Salesforce 2 years after founding it for $76M. Incredible.)\n
• #86: (With a quick shout-out to my co-author, Ben Yoskovitz. He credits his entire career to blogging; he even has an infographic to prove it. Ben, Jevon, and the GoInstant team sold their 14-person company to Salesforce 2 years after founding it for $76M. Incredible.)\n
• #87: (With a quick shout-out to my co-author, Ben Yoskovitz. He credits his entire career to blogging; he even has an infographic to prove it. Ben, Jevon, and the GoInstant team sold their 14-person company to Salesforce 2 years after founding it for $76M. Incredible.)\n
• #88: Jeff Jarvis does a great job of explaining thrust/scarcity, versus lube/abundance thinking when he compares traditional publishers to companies like Facebook and Google.\n
• #89: When the system is frictionless, other good things happen—demand grows dramatically.\n
• #90: Lube companies make things frictionless, then reap the rewards because of that abundance.\n
• #91: If you’re a startup challenging corporate giants—and I assume most of you are—then your future is fighting thrust-thinking companies with lube-inspired strategies.\n
• #92: Fighting thrust with thrust is positively Medieval.\n
• #93: But fighting thrust with lube is sexy.\n
• #94: For one thing, it’s easy to start stuff.\n
• #95: Consider the costs of getting into a business, then and now.\n
• #96: Consider the costs of getting into a business, then and now.\n
• #97: Consider the costs of getting into a business, then and now.\n
• #98: Consider the costs of getting into a business, then and now.\n
• #99: Consider the costs of getting into a business, then and now.\n
• #100: Consider the costs of getting into a business, then and now.\n
• #101: Upfront costs are almost nothing these days.\n
• #102: The old, thrust-centric method is ready, aim, fire. Get your business plan, get financing, and launch.\n
• #103: Instead, try this: First, shoot wildly in all directions. When something goes “ow” keep shooting that way until you hit it reliably. Then go get money. \n
• #104: You can do this because, thanks to the frictionless world of electrons, the bullets are free.\n
• #105: That means you’re always learning, too.\n
• #106: When we launched Lean Analytics last week, we ran experiments. We know which proponents are most likely to have followers that complete a survey, or followers that click on the picture of the book. And we can use that.\n
• #107: In fact, pretty much everything is up for analysis. And doing so is free—because it’s frictionless.\n
• #108: Sure, today most of that tracking is on the web. But in an augmented, mobile, connected world, more and more things are trackable.\n
• #109: Soon, we’ll be so tied to our online worlds, we’ll be trackable too.\n
• #110: And with frictionless data and ubiquitous collection, everything is an experiment\n
• #111: Sometimes these experiments come from the weirdest places, and have counter-intuitive results. Like this study by NYU professor Panos Ipierotis that concluded that how well a review is written beats what the reviewer game. This led Zappos to correct review spelling, which drove sales up.\n
• #112: And this, in turn, means embracing your inner cyborg.\n
• #113: All of this data is hard to make sense of when we’re trying to compete and innovate at scale.\n
• #114: As it turns out, the answer is combining humans and machines.\n
• #115: Lube startups need to be data-informed, without becoming slaves to the data. That’s because while algorithms and machines can optimize to something called a Local Maxima, they can’t redesign something.\n
• #116: This is a fundamental problem with evolutionary optimization. To illustrate, let me tell you why you’ll always have a blind spot. There’s a place in your eye where you can’t see things. The other eye compensates, and your brain fills in the rest, but you really can’t see part of what you look at.\n
• #117: The eye has evolved in nature in many, many different forms, often independently. Our eyes and those of cephalopods like the octopus look similar, but have very different histories. The human eye is “backwards”—the infrastructure is actually in front of the retina. That means the optic nerve runs in front of the part that catches the light. This is because the human eye evolved as a growth of the brain. On the other hand, the octopus eye has the optic nerve behind the retina, because it evolved as a dent in the head. Which means octopuses don’t have a blind spot. But you’re not going to evolve a better eye, because the intermediate mutations would be worse.\n
• #118: The eye has evolved in nature in many, many different forms, often independently. Our eyes and those of cephalopods like the octopus look similar, but have very different histories. The human eye is “backwards”—the infrastructure is actually in front of the retina. That means the optic nerve runs in front of the part that catches the light. This is because the human eye evolved as a growth of the brain. On the other hand, the octopus eye has the optic nerve behind the retina, because it evolved as a dent in the head. Which means octopuses don’t have a blind spot. But you’re not going to evolve a better eye, because the intermediate mutations would be worse.\n
• #119: The eye has evolved in nature in many, many different forms, often independently. Our eyes and those of cephalopods like the octopus look similar, but have very different histories. The human eye is “backwards”—the infrastructure is actually in front of the retina. That means the optic nerve runs in front of the part that catches the light. This is because the human eye evolved as a growth of the brain. On the other hand, the octopus eye has the optic nerve behind the retina, because it evolved as a dent in the head. Which means octopuses don’t have a blind spot. But you’re not going to evolve a better eye, because the intermediate mutations would be worse.\n
• #120: The eye has evolved in nature in many, many different forms, often independently. Our eyes and those of cephalopods like the octopus look similar, but have very different histories. The human eye is “backwards”—the infrastructure is actually in front of the retina. That means the optic nerve runs in front of the part that catches the light. This is because the human eye evolved as a growth of the brain. On the other hand, the octopus eye has the optic nerve behind the retina, because it evolved as a dent in the head. Which means octopuses don’t have a blind spot. But you’re not going to evolve a better eye, because the intermediate mutations would be worse.\n
• #121: Imagine I gave an algorithm three wheels and asked it to optimize their configuration. It would arrive at some form of tricycle. But it wouldn’t go, “hey, can I get a fourth wheel?” That would be what a human would do. That’s a leap of faith.\n
• #122: Imagine I gave an algorithm three wheels and asked it to optimize their configuration. It would arrive at some form of tricycle. But it wouldn’t go, “hey, can I get a fourth wheel?” That would be what a human would do. That’s a leap of faith.\n
• #123: Imagine I gave an algorithm three wheels and asked it to optimize their configuration. It would arrive at some form of tricycle. But it wouldn’t go, “hey, can I get a fourth wheel?” That would be what a human would do. That’s a leap of faith.\n
• #124: Imagine I gave an algorithm three wheels and asked it to optimize their configuration. It would arrive at some form of tricycle. But it wouldn’t go, “hey, can I get a fourth wheel?” That would be what a human would do. That’s a leap of faith.\n
• #125: Imagine I gave an algorithm three wheels and asked it to optimize their configuration. It would arrive at some form of tricycle. But it wouldn’t go, “hey, can I get a fourth wheel?” That would be what a human would do. That’s a leap of faith.\n
• #126: Imagine I gave an algorithm three wheels and asked it to optimize their configuration. It would arrive at some form of tricycle. But it wouldn’t go, “hey, can I get a fourth wheel?” That would be what a human would do. That’s a leap of faith.\n
• #127: Imagine I gave an algorithm three wheels and asked it to optimize their configuration. It would arrive at some form of tricycle. But it wouldn’t go, “hey, can I get a fourth wheel?” That would be what a human would do. That’s a leap of faith.\n
• #128: Imagine I gave an algorithm three wheels and asked it to optimize their configuration. It would arrive at some form of tricycle. But it wouldn’t go, “hey, can I get a fourth wheel?” That would be what a human would do. That’s a leap of faith.\n
• #129: This is why the best ways to optimize things and make sense of all the data blend humans and machines. Reddit, for example, is a cyborg that surfaces the best things on the Internet. It may not look like the T-1000, but it’s a cyborg.\n
• #130: With tools like Mechanical Turk, building these kinds of human/machine hybrid is trivially easy and cheap.\n
• #131: Okay, that’s been a bit fanciful. So let me talk about three concrete examples.\n
• #132: First: taxis. There are 13,087 taxis in the city with medallions. That’s $6.5 Billion dollars in taxi medallion speculation. And what do you get for it? The right to pick up and drop off without pre-arrangement. Basically, the taxi world handles the problem with thrust thinking—dispatches and lobbying.\n
• #133: Uber is a startup that’s trying to change that. In fact, they just won a big victory in DC yesterday when a city amendment that would force them to charge three times as much as taxis was struck down. Protectionism and lobbying failed.\n
• #134: But that’s not why Uber is cool. They make money BECAUSE it’s easy to connect riders and drivers. But they do it better by analyzing the data, so they tell drivers when and where to go in order to maximize fares, based on what they learn about usage patterns.\n
• #135: Here’s a second example: flipping the classroom. In 2004 Jonathan Bergmann and Aaron Sams decided to change the way they taught. They made homework schoolwork and vice versa.\n
• #136: They realized that once, the role of teaching was to broadcast knowledge—to find the best person in the village to teach. But that now, broadcast was easy, because of digital technology; making knowledge stick was hard.\n
• #137: If you can get the best teacher in the world into the classroom, doesn’t it make sense to make that the homework, and make the working out of problems the classroom part? Bergmann and Sams realized that a frictionless classroom should be flipped—and students reaped the benefits.\n
• #138: Finally, consider manufacturing, and the Makerbot.\n
• #139: These things eat extruded filament or corn starch\n
• #140: In a world where we can print anything we want, what happens to Toys R Us?\n
• #141: Then what happens to UPS?\n
• #142: And what happens to the entire supply chain of automotive parts mandated by Lemon Laws?\n
• #143: When Canadian Tire has a lathe at each store?\n
• #144: So, in conclusion\n
• #145: The next fifty big companies will be those that figure out where an established market leader relies on thrust, instead of lube, to address the inherent friction in their marketplace.\n
• #146: And that’s why this is the best time of all to be a disruptor. Because we’re moving from atoms to bits.\n
• #147: We just have to make money from because\n
• #148: Fire first and scale later\n
• #149: and act like everything is an experiment.\n
• #150: Build your companies for the day when friction is gone\n
• #151: And enjoy the lube.\n
• #152: Thanks. And remember—we’d love it if you’d visit this URL, sign up and join the conversation.\n