Why Information Grows

It is hard for us humans to separate information from meaning because we cannot help interpreting messages. We infuse messages with meaning automatically, fooling ourselves to believe that the meaning of a message is carried in the message. But it is not. This is only an illusion. Meaning is derived from context and prior knowledge. Meaning is the interpretation that a knowledge agent, such as a human, gives to a message, but it is different from the physical order that carries the message, and different from the message itself. Meaning emerges when a message reaches a life-form or a machine with the ability to process information; it is not carried in the blots of ink, sound waves, beams of light, or electric pulses that transmit information.
 

From the book Why Information Grows by Cesar Hidalgo. I read this book a long ways back in 2017,  but it's of no less interest now.

And it is the arrow of complexity—the growth of information—that marks the history of our universe and species. Billions of years ago, soon after the Big Bang, our universe did not have the capacity to generate the order that made Boltzmann marvel and which we all take for granted. Since then, our universe has been marching toward disorder, as Boltzmann predicted, but it has also been busy producing pockets that concentrate enormous quantities of physical order, or information. Our planet is a chief example of such a pocket.
 

When one first encounters the second law of thermodynamics, it's easy to tumble into despair at the pointlessness of everything. With the universe fated to collapse into heat death eventually, what is the point of it all?

In this existential void, the presence of pockets of information and order can feel like symbols of rebellion, a raised fist spray painted on a fragment of wall that remains from a bombed-out building. In manifestations of order we see intent, in intent we interpret meaning, and in meaning we find comfort.

Information, when understood in its broad meaning as physical order, is what our economy produces. It is the only thing we produce, whether we are biological cells or manufacturing plants. This is because information is not restricted to messages. It is inherent in all the physical objects we produce: bicycles, buildings, streetlamps, blenders, hair dryers, shoes, chandeliers, harvesting machines, and underwear are all made of information. This is not because they are made of ideas but because they embody physical order. Our world is pregnant with information. It is not an amorphous soup of atoms, but a neatly organized collection of structures, shapes, colors, and correlations. Such ordered structures are the manifestations of information, even when these chunks of physical order lack any meaning.
 

There are plenty of books on information theory, and viewing the universe through the lens of information and computation is increasingly popular, but Hidalgo's book is more readable than most.

To battle disorder and allow information to grow, our universe has a few tricks up its sleeve. These tricks involve out-of-equilibrium systems, the accumulation of information in solids, and the ability of matter to compute.
 
It is the growth of information that unifies the emergence of life with the growth of economies, and the emergence of complexity with the origins of wealth.
 
In twenty-six minutes Iris traveled from the ancientness of her mother’s womb to the modernity of twenty-first-century society. Birth is, in essence, time travel.
 

Birth as time travel is one of those metaphors that, once heard, lodges in your mind like something you always knew. When Arnold Schwarzenegger time travels back from the future to the modern day in The Terminator, he arrives naked, like a newborn.

[It is unclear why a cyborg from the future speaks with a thick Austrian accent, one of the only mysteries I have always hoped would be explained in some throwaway expository joke. My guess is that the voice was a marketing Easter Egg, like celebrity voices in Waze, and someone forgot to flip the Terminator back to its factory default voice before sending it back in time.]

Humans are special animals when it comes to information, because unlike other species, we have developed an enormous ability to encode large volumes of information outside our bodies. Naively, we can think of this information as the information we encode in books, sheet music, audio recordings, and video. Yet for longer than we have been able to write we have been embodying information in artifacts or objects, from arrows to microwave ovens, from stone axes to the physical Internet. So our ability to produce chairs, computers, tablecloths, and wineglasses is a simple answer to the eternal question: what is the difference between us, humans, and all other species? The answer is that we are able to create physical instantiations of the objects we imagine, while other species are stuck with nature’s inventory.
 

Another reason humans wouldn't evolve on a gaseous planet like Jupiter, besides the fact that we'd just burn up, is that without any solids we'd have no way of encoding information to pass on to future generations. Therefore, any advanced civilization in the universe would, it would seem, live in physical conditions that allow for the formation of solids, but not solids that are too rigid.

The temperature band matters. We need solids that are malleable to encode richer sets of information. Add to that the ability to compute, which we see in all forms in our world, down to the cellular level, and suddenly you have life. There is logic to why we look for specific conditions in the universe as precursors for life, and it can be defined more broadly than just looking for water, which is a downstream condition. Further upstream we just want a planet with solids, in a particular band of temperatures.

Such conditions allow living creatures to record and pass along information to the next generation. When humans finally were able to do so, they in effect conquered time. No longer did the knowledge of one generation evaporate into the sinkhole of mortality.

The car’s dollar value evaporated in the crash not because the crash destroyed the atoms that made up the Bugatti but because the crash changed the way in which these were arranged. As the parts that made the Bugatti were pulled apart and twisted, the information that was embodied in the Bugatti was largely destroyed. This is another way of saying that the $2.5 million worth of value was stored not in the car’s atoms but in the way those atoms were arranged. That arrangement is information.
 
...
 
So the value of the Bugatti is connected to physical order, which is information, even though people still debate what information is. According to Claude Shannon, the father of information theory, information is a measure of the minimum volume of communication required to uniquely specify a message. That is, it’s the number of bits we need to communicate an arrangement, like the arrangement of atoms that made the Bugatti.
 
...
 
The group of Bugattis in perfect shape, however, is relatively small, meaning that in the set of all possible rearrangement of atoms—like people moving in a stadium—very few of these involve a Bugatti in perfect condition. The group of Bugatti wrecks, on the other hand, is a configuration with a higher multiplicity of states (higher entropy), and hence a configuration that embodies less information (even though each of these states requires more bits to be communicated). Yet the largest group of all, the one that is equivalent to people sitting randomly in the stadium, is the one describing Bugattis in their “natural” state. This is the state where iron is a mineral ore and aluminum is embedded in bauxite. The destruction of the Bugatti, therefore, is the destruction of information. The creation of the Bugatti, on the other hand, is the embodiment of information.
 

One can separate out the intrinsic value of an item, defined above as the rarity of the state of the configuration of that item, from the external value of an item, as defined by qualities such as symbolic or emotional ones, like nostalgia.

In Pulp Fiction, Bruce Willis risks life and limb to recover a watch given to him by his father. There's no evidence it's a particularly rare watch, he could likely buy another just like it, but its symbolic value to him is extrinsic to the item yet tethered to it the way a genie is trapped in a magic lantern (and that special meaning is conveyed in the now immortal speech by Christopher Walken).

Even the most rational people I know own something that's not physically rare but emotionally rich, a talisman or totem that they use to summon whatever power it holds, whether it be nostalgia or regret or some other enchantment known only to themselves.

What Shannon teaches us is that the amount of information that is embodied in a tweet is equal to the minimum number of yes-or-no questions that Brian needs to ask to guess Abby’s tweet with 100 percent accuracy. But how many questions is that?
 
Shannon’s theory tells us that we need 700 bits, or yes-or-no questions, to communicate a tweet written using a thirty-two-character alphabet. Shannon’s theory is also the basis of modern communication systems.
 

One mathematical reason for the rising usage of emoji in Twitter and other forms of online communication may be that it increases the amount of information that can be encoded in 140 (and now 280) characters.

You'll recall from earlier that the third of the three conditions that allow information to grow is the ability of matter to compute.

To illustrate the prebiotic nature of the ability of matter to process information, we need to consider a more fundamental system. Here is where the chemical systems that fascinated Prigogine come in handy. Consider a set of chemical reactions that takes a set of compounds {I} and transforms them into a set of outputs {O} via a set of intermediate compounds {M}. Now consider feeding this system with a steady flow of {I}. If the flow of {I} is small, then the system will settle into a steady state where the intermediate inputs {M} will be produced and consumed in such a way that their numbers do not fluctuate much. The system will reach a state of equilibrium. In most chemical systems, however, once we crank up the flow of {I} this equilibrium will become unstable, meaning that the steady state of the system will be replaced by two or more stable steady states that are different from the original state of equilibrium.13 When these new steady states emerge, the system will need to “choose” among them, meaning that it will have to move to one or the other, breaking the symmetry of the system and developing a history that is marked by those choices. If we crank up the inflow of the input compounds {I} even further, these new steady states will become unstable and additional new steady states will emerge. This multiplication of steady states can lead these chemical reactions to highly organized states, such as those exhibited by molecular clocks, which are chemical oscillators, compounds that change periodically from one type to another. But does such a simple chemical system have the ability to process information? Now consider that we can push the system to one of these steady states by changing the concentration of inputs {I}. Such a system will be “computing,” since it will be generating outputs that are conditional on the inputs it is ingesting. It would be a chemical transistor. In an awfully crude way this chemical system models a primitive metabolism. In an even cruder way, it is a model of a cell differentiating from one cell type to another—the cell types can be viewed abstractly as the dynamic steady states of these systems, as the complex systems biologist Stuart Kauffman suggested decades ago. Highly interacting out-of-equilibrium systems, whether they are trees reacting to the change of seasons or chemical systems processing information about the inputs they receive, teach us that matter can compute. These systems tell us that computation precedes the origins of life just as much as information does. The chemical changes encoded by these systems are modifying the information encoded in these chemical compounds, and therefore they represent a fundamental form of computation. Life is a consequence of the ability of matter to compute.
 

What's lovely about all of these conditions that allow information to grow is their seeming relevance to individuals and groups of individuals, like corporations or societies or markets.

Humans are concentrated bundles of information with compute power, and when we push ourselves out of equilibrium, we accumulate information. When we crank up our inputs and force ourselves out of our own equilibrium, as we do when we become students, we grow as we restore ourselves to steady state. Whenever anyone complains that they're in a rut, I always counsel them to force themselves out of equilibrium.

***

That covers much of the first half of the book, all fascinating. However, the part of the book that's of broader interest to a business audience is Hidalgo's discussion of the economy as a creator of information.

It's easiest to understand the information creation capacity of an economy by examining its outputs, and the simplest outputs to understand are physical products.

Thinking about products as crystals of imagination tells us that products do not just embody information but also imagination. This is information that we have generated through mental computations and then disembodied by creating an object that mimics the one we had in our head. Edible apples existed before we had a name for them, a price for them, or a market for them. They were present in the world. As a concept, apples were simply imported into our minds. On the other hand, iPhones and iPads are mental exports rather than imports, since they are products that were begotten in our minds before they became part of our world. So the main difference between apples and Apples resides in the source of their physical order rather than in their embodiment of physical order. Both products are packets of information, but only one of them is a crystal of imagination.
 

Like many navel gazers in the tech industry, I'm guilty of stereotyping companies. Apple's strength is integrated hardware and software, Google is the king of machine learning and crunching large data sets, Facebook is the social network to end all social networks, and Amazon is the everything platform.

However, if you haven't worked or been inside any of those companies, it's fairest to judge them as black boxes into which inputs disappear and come out as various outputs, usually products and services like gadgets or websites or applications. Everything else is a mild form of fan fiction. 

By analyzing a company's outputs, one can deduce a great deal about its capabilities. Hidalgo does the same but at the country level.

The idea of crystallized imagination tells us that a country’s export structure carries information about more than just its abundance of capital and labor. A country’s export structure is a fingerprint that tells us about the ability of people in that country to create tangible instantiations of imaginary objects, such as automobiles, espresso machines, subway cars, and motorcycles, and of course about the myriad of specific factors that are needed to create these sophisticated products. In fact, the composition of a country’s exports informs us about the knowledge and knowhow that are embodied in that country’s population.
 

A country that can export a product like an iPhone generally has greater generative power than one that can only export raw materials like bananas. The telltale clues to the economic potential of a country lie not in its imports but its exports.

So what has any of this to do with Chile? The only connection between Chile and the history of electricity comes from the fact that the Atacama Desert is full of copper atoms, which, just like most Chileans, were utterly unaware of the electric dreams that powered the passion of Faraday and Tesla. As the inventions that made these atoms valuable were created, Chile retained the right to hold many of these atoms hostage. Now Chile can make a living out of them. This brings us back to the narrative of exploitation we described earlier. The idea of crystallized imagination should make it clear that Chile is the one exploiting the imagination of Faraday, Tesla, and others, since it was the inventors’ imagination that endowed copper atoms with economic value. But Chile is not the only country that exploits foreign creativity this way. Oil producers like Venezuela and Russia exploit the imagination of Henry Ford, Rudolf Diesel, Gottlieb Daimler, Nicolas Carnot, James Watt, and James Joule by being involved in the commerce of a dark gelatinous goo that was virtually useless until combustion engines were invented. Making a strong distinction between the generation of value and the appropriation of monetary compensation helps us understand the difference between wealth and economic development. In fact, the world has many countries that are rich but still have underdeveloped economies. This is a distinction that we will explore in detail in Part IV. But making this distinction, which comes directly from the idea of crystallized imagination, helps us see that economic development is based not on the ability of a pocket of the economy to consume but on the ability of people to turn their dreams into reality. Economic development is not the ability to buy but the ability to make.
 

At a corporate level, I can recall an age when Sony was the king of consumer electronics the world over. I first coveted a Walkman, then later a Discman. Our family spent its formative years huddled around a giant (at the time) Sony Trinitron TV, and we were the envy of all my friends for owning one. I looked forward to any trip to Japan for a chance to walk the electronics districts to purchase the coolest gadgets on the planet, and for years I owned a Minidisc player model that you couldn't find in the U.S.

And then the world shifted, and the gadget which subsumed all other gadgets was the computer, and as it shrank in size while growing in computational power, the way we interacted with such devices increasingly became software-based. In that competition, the vector which mattered more than anything became software design, a skill Sony had not mastered.

The company that understood both software and hardware design better than any company in the world happened to be located in Silicon Valley, not Japan, and, after a long Wintel interregnum, caused by a number of business factors covered comprehensively elsewhere, Apple's unique skills found themselves in a universe they could really dent. And dent they did.

Thanks especially to the market opportunity created by the smartphone, which it seized with the iPhone, Apple not only surpassed Sony and moved the balance of power in consumer technology across the Pacific Ocean to American shores but became the most valuable company in the entire world.

***

Not all information is easily embodied. For example, for a while I puzzled over what I'll call the Din Tai Fung Paradox.

Din Tai Fung is a restaurant chain, and I visited the original outlet in Taipei decades ago with my mother. They're known for their Shanghainese soup dumplings, made with a very delicate wrap that somehow never breaks and dumps its precious cargo of pork broth until the moment at which you prod it with your chopsticks just so. Some will argue whether Ding Tai Fung is all that and a bucket of chicken, but at a minimum I find the menu to be satisfying comfort food done consistently, in a setting that is usually cleaner and more well-kept than your average chain restaurant outlet. You'll find superior deals from a street vendor and more elaborate preparation at a higher-end restaurant, but Din Tai Fung industrializes and scales a Chinese staple. We don't pay enough attention to scale.

The mystery is why Din Tai Fung has opened so few outlets; they've only dropped locations in a handful of cities in about ten countries in the world, and every Din Tai Fung is packed solid from open to close with the type of ever-present line of humans snaked outside the front door that you so rarely see at any restaurant, let alone a chain.

For a few months, a new outlet was rumored to be opening in San Francisco soon, and among my friends it was as momentous a rumor as if a new Star Wars teaser trailer had dropped. Ultimately, one opened in the Bay Area, but in Santa Clara instead of San Francisco. 

Which leads to a further mystery: why haven't any competing chains opened up to make the same items to fill the market void? I would never open a restaurant, but my family knows I'd make an exception if I were granted the opportunity to open a branch of Din Tai Fung anywhere. I bring it up every family gathering, when there's a lull in the conversation. Forget cryptocurrency, I want to mint me some Din Tai Fung coin.

At every Din Tai Fung I've been to, they have a glass window so you can look into the kitchen to see the soup dumplings being wrapped, always by kitchen staff wearing white uniforms, almost like lab assistants, an impression magnified by the branches that require face masks. It's rumored that the branches in Asia try to hire the tallest, most attractive men to man the soup dumpling assembly line, but it sounds about as true as a lot of things my aunts and uncles tell me, which is to say it's more credible than I'd care to admit.

The hermetic vibe behind the glass is as far from the vendor selling goods from a street cart as possible; some find street food charming but if you're taking this food to a global audience it needs to be sanitized or sterilized, the same way movies for the Chinese market strip out any storylines that might offend. It's not just the front of the house that's immaculate, the show behind the glass display says they have nothing to hide. It's the equivalent of the blackjack dealer at a casino clapping and turning their hands one way and the other before moving to the next table.

More interesting to me was that Din Tai Fung even doesn't even bother to hide the process behind its staple dish, the evidence is on thousands of smart phones by this point, everyone seems to stop to take a photo or video of the assembly line while waiting for their table.

And yet any Chinese food fan knows it's notoriously hard to find a good soup dumpling. In this age where recipes for almost anything are available online for free, why can't you find a good soup dumpling in most major cities in the world? Or, for that matter, a good burrito, or any dish you love? Why are these crystals of imagination so unevenly distributed when the recipes for making them so broadly available?

The answer, as any home chef who has tried to make a dish from some highfalutin cookbook knows, is simple: you can have the most precise ingredient list and directions and still struggle to make anything approaching what you ate, whether it came from a $400 tasting menu or a mainstream cookbook. Cooking is not nearly as deterministic as the term recipe implies.

Slight variations in environment, weather, ingredients, and cookware can lead to massive differences in the final product. Your oven may say 400 degrees, but the actual temperature inside, at the precise spot where you've placed your baking dish, may be different. That celery you use for your mirepoix today may not be as fresh as the celery you used last week. The air pressure where you're cooking on a particular day may differ from that where you live, the bacteria in the air may also vary. Great chefs appear on Top Chef and flail making dishes they've made hundreds of times in their own restaurant kitchen because every bit of environmental variation matters.

We may glamorize the image of the genius, heroic chef, working magic to create a delicious and beautifully plated dish that a waiter places before us with a balletic flourish, but the true value creation in a restaurant comes from translating that moment of genius into a rote, repeatable cycle. The popularity of sous vide as a cooking technique, even at high end restaurants, comes down its repeatable precision and accuracy. Ask any chef and they'll tell you the value of a line chef who can cook dozens of proteins to the right level of doneness every time given the high cost of fish and meat.

In addition to all those conditionals, much of cooking skill comes down to learned muscle memory and pattern recognition that can only be encoded in a human being through repeated trial and error. I tried to learn some of my favorite of my mother and grandmother's dishes by writing down recipes they dictated to me, but much was lost in translation. Like so much maternal magic, it could only be learned, truly, at their side, with an apron on, watching, imitating, botching one dish after another, until some of it seeped into my bones.

In a memorable segment from the documentary Jiro Dreams of Sushi, apprentice Daisuke Nakazawa is assigned the job of making egg sushi, or tamago. He believes it will be simple, but again and again, Jiro rejects his work. Nakazawa ends up making over 200 rejected samples until finally, one day, Jiro approves. Nakazawa cries in relief and joy.

***

Hardware and software are not like cooking. When knowledge and instructions can be encoded in bits, a level of precision is possible that is effectively, for the purposes of this discussion, deterministic. Manufacturing a hundred million iPhones is like food production, but not the type done in high end or home kitchens. Instead, it is more like producing a hundred million Oreos.

There is one country in the world where that many iPhones can be manufactured for the cost that allows Apple to reap its insane profits: China. I can't think of any other country in the world, not India or Mexico or the United States, or all of Europe together that can make that many iPhones for that price to meet the market demand year after year. Some countries have the labor but not the skills, others have the skills but not enough labor, and others just can't do the work as cheaply as China can.

Recall that the potential of an economy can be judged by the complexity of its exports. Based on that, it's difficult to imagine an economy outside of the U.S. with more potential than China. Some of the most complex products in the world, and the iPhone deserves to be on that list, are made in China.

I've backed many a Kickstarter hardware project, and without fail, every one has been made in China, usually Shenzhen. Inevitably, when the products are delayed, the project's creators will send an update with some photos of a few of them in China, at some plant, examining some part that will get the project back on track, or with their arms around a few Chinese plant managers giving a thumbs up sign.

Kickstarter often feels like an industrial and software design and marketing operation layer grafted on top of the manufacturing capabilities of Shenzhen. It is an early warning indicator of China's economic potential, and the gap that remains to realizing it.

Here is another. Foxconn assembles iPhones for Apple, and for their efforts they make anywhere from $8 to $30 per iPhone, depending on what article you believe. Whatever the figure, we know it is not far off from that.

Apple, in contrast, makes hundreds of dollars per iPhone. They earn that premium, many multiples what Foxconn earns, by virtue of being the ones who designed every aspect of the phone, from the software to the hardware. China can supply labor and even sometimes components, but the crystal of imagination that is the iPhone, perhaps the most valuable such crystal in the history of the world, comes almost entirely from the imagination of employees of Apple. Foxconn is one cog in a long supply chain, and that link isn't the one made of gold.

However, to even have the capability of making an iPhone for less than the cost of a lunch in San Francisco is a skill, one China has shown again and again. Many another country wishes it had such a demonstrated skill. Were China ever able to gain some of the software and industrial design skills of a company like Apple, they would be even more of an economic powerhouse than they are now.

That's a massive conditional. It's not something that can be learned by mere handwaving or even sheer industriousness. After all, Sony could return to its former glory, or Samsung be even more dominant globally, if software design skills were so easily learned.

Someone someday will write a book of the history of software design and how it came to be that Apple built up that capability more than any other technology company, and I'll be among its most eager readers because it's an untold story that holds the key to one of the greatest value creation stories in the history of business.

...our world is one in which knowledge and knowhow are “heavier” than the atoms we use to embody their practical uses. Information can be moved around easily in the products that contain it, whether these are objects, books, or webpages, but knowledge and knowhow are trapped in the bodies of people and the networks that these people form. Knowledge and knowhow are so “heavy” that when it comes to a simple product such as a cellphone battery, it is infinitely easier to bring the lithium atoms that lie dormant in the Atacama Desert to Korea than to bring the knowledge of lithium batteries that resides in Korean scientists to the bodies of the miners who populate the Atacaman cities of Antofagasta and Calama. Our world is marked by great international differences in countries’ ability to crystallize imagination. These differences emerge because countries differ in the knowledge and knowhow that are embodied in their populations, and because accumulating knowledge and knowhow in people is difficult. But why is it hard for us to accumulate the knowledge and knowhow we need to transform our dreams into reality?
 

If knowledge were so easy to transfer, I'd be a three-star Michelin Chef because someone gifted me a copy of the Eleven Madison Park cookbook.

Getting knowledge inside a human’s nervous system is not easy because learning is both experiential and social.5 To say that learning is social means that people learn from people: children learn from their parents and employees learn from their coworkers (I hope). The social nature of learning makes the accumulation of knowledge and knowhow geographically biased. People learn from people, and it is easier for people to learn from others who are experienced in the tasks they want to learn than from people with no relevant experience in that task. For instance, it is difficult to become an air traffic controller without learning the trade from other air traffic controllers, just as it is difficult to become a surgeon without having ever been an intern or a resident at a hospital. By the same token, it is hard to accumulate the knowhow needed to manufacture rubber tires or an electric circuit without interacting with people who have made tires or circuits.6 Ultimately, the experiential and social nature of learning not only limits the knowledge and knowhow that individuals can achieve but also biases the accumulation of knowledge and knowhow toward what is already available in the places where these individuals reside. This implies that the accumulation of knowledge and knowhow is geographically biased.
 

What governs the information production capacity of a country? Hidalgo coins two terms to analyze this problem. One is the personbyte.

We can simplify this discussion by defining the maximum amount of knowledge and knowhow that a human nervous system can accumulate as a fundamental unit of measurement. We call this unit a personbyte, and define it as the maximum knowledge and knowhow carrying capacity of a human.
 

The other term is firmbyte.

The limited proliferation of megafactories like the Rouge implies that there must be mechanisms that limit the size of the networks we call firms and make it preferable to disaggregate production into networks of firms. This also suggests the existence of a second quantization limit, which we will call the firmbyte. It is analogous to the personbyte, but instead of requiring the distribution of knowledge and knowhow among people, it requires them to be distributed among a network of firms.
 

Hidalgo then delves a bit into Coase's transaction cost theory of the firm. Traditionally, Coase's theory is used as a way to explain why firms are fundamentally limited in their size, the idea being that at some size, external transactions become cheaper than internal coordination costs and so it's more efficient to just transact externally rather than produce internally.

I'm not interested in examining that topic now. Instead, let's assume that firms all do have some asymptote in size beyond which Coase's anchor becomes too heavy. The interesting implication is that given the existence of a ceiling on the size of the firmbyte, if some chunk of knowledge exceeds that capacity then it can only be carried by a network of firms.

It's long been said that the center of the technology universe shifted from Boston's route 128 to Silicon Valley because California banned non-competes (here's one study). Hidalgo's theory of the finite compute ability of a network of humans and firms explains how this works. The free movement of employees in Silicon Valley allows the region's knowledge-carrying capacity to increase at the expense of any single firm's benefit. Per Coase, the cost of information movement in Silicon Valley, as embodied by an employee carrying a personbyte from one firm to the next, is lower than it was in the route 128 corridor.

Let's telescope back out to the country level. What applies at the regional or industry level holds at the country level. A country's knowledge carrying capacity, and thus its information production power, is influenced in part by the size of networks it can form.

In his 1995 book Trust, he [Francis Fukuyama] argues that the ability of a society to form large networks is largely a reflection of that society’s level of trust. Fukuyama makes a strong distinction between what he calls “familial” societies, like those of southern Europe and Latin America, and “high-trust” societies, like those of Germany, the United States, and Japan.
 
Familial societies are societies where people don’t trust strangers but do trust deeply the individuals in their own families (the Italian Mafia being a cartoon example of a familial society). In familial societies family networks are the dominant form of social organization where economic activity is embedded, and are therefore societies where businesses are more likely to be ventures among relatives. By contrast, in high-trust societies people don’t have a strong preference for trusting their kin and are more likely to develop firms that are professionally run. Familial societies and high-trust societies differ not only in the composition of the networks they form—as in kin and non-kin—but also in the size of the networks they can form. This is because the professionally run businesses that evolve in high-trust societies are more likely to result in networks of all sizes, including large ones. In contrast, familial societies are characterized by a large number of small businesses and a few dominant families controlling a few large conglomerates.
 
Yet, as we have argued before, the size of networks matters, since it helps determine the economic activities that take place in a location. Larger networks are needed to produce products of higher complexity and, in turn, for societies to achieve higher levels of prosperity. So according to Fukuyama, the presence of industries of different sizes indicates the presence of trust. In his own words: “Industrial structure tells an intriguing story about a country’s culture. Societies that have very strong families but relatively weak bonds of trust among people unrelated to one another will tend to be dominated by small, family-owned and managed business. On the other hand, countries that have vigorous private nonprofit organizations like schools, hospitals, churches, and charities, are also likely to develop strong private economic institutions that go beyond the family.”
 

In Tyler Cowen's conversation with economist Luigi Zingales, the latter hints at the limitations of familial economies in humorous fashion:

One friend of mine was saying that the demise of the Italian firm family structure is the demise of the Italian family. In essence, when you used to have seven kids, one out of seven in the family was smart. You could find him. You could transfer the business within the family with a little bit of meritocracy and selection.
 
When you’re down to one or two kids, the chance that one is an idiot is pretty large. The result is that you can’t really transfer the business within the family. The biggest problem of Italy is actually fertility, in my view, because we don’t have enough kids. If you don’t have enough kids, you don’t have enough people to transfer. You don’t have enough young people to be dynamic.
 
The Italian culture has a lot of defects, but the entrepreneurship culture was there, has been there, and it still is there, but we don’t have enough young people.
 

Low fertility's impact on economies is an issue globally, for example in Japan, but low trust outside of family is an even broader constraint on the knowledge carrying capacity of an economy. If you can't form as large a firm as another country, you can't compete in some businesses and the information producing capability of your economy has a lower ceiling.

If you run a company, you're no doubt familiar with the efficiency gains that arise when different employees and departments operate with high trust. Links form easily given an assumption of low risk, and knowledge moves more quickly, fluidly. Networks then facilitate trust in a virtuous cycle, an example being the military as an integrating institution in a multi-ethnic society.

Trust based on family has its own advantages, but for now I'm focused on an economy's ceiling, and networks that throw off the shackles of family-based firms can scale more. China not only has the population to supply a workforce that can assemble 100 million iPhones in a year, it has an economy that has moved beyond any roots in family-based trust.

Hidalgo's theory also explains why we don't see geographic leakage in industry know-how. Why aren't there Silicon Valleys everywhere?

The personbyte theory can also help us explain why large chunks of knowledge and knowhow are hard to accumulate and transfer, and why knowledge and knowhow are organized in the hierarchical pattern that is expressed in the nestedness of the industry-location data. This is because large chunks of knowledge and knowhow need large networks of people to be embodied in, and transferring or duplicating large networks is not as easy as transferring a small group of people. As a result, industry-location networks are nested, and countries move to products that are close by in the product space.
 

When the knowledge required to create something like an iPhone or a Hollywood film require the interaction of multiple people, with all their accumulated knowledge, seizing it for yourself isn't as easy as poaching one employee or sprinting off with a burning branch to give fire to mankind like Prometheus. Thus we understand why, besides its weather, LA has such a grip on filmmaking for the global market, why any handset manufacturers can't just reverse engineer an iPhone, and why, despite having hundreds of millions of users for iMessages, Apple isn't a credible threat in social networking.

When I study the Chinese tech market, I see an incredibly high ceiling. In fact, the Chinese consumer market in tech is more dynamic now than its counterpart in Silicon Valley. Once, China was belittled for simply copying all the US tech companies. It's true, there is a Chinese Bizarro instance of all successful U.S. tech companies, a Chinese Google, Facebook, Amazon, Twitter, Instagram, YouTube, and so on.

Thanks to that complex interaction of culture and technology, however, China now creates companies with no real American equivalents, and that extends beyond WeChat. China also has more dense cities than America, and density creates its own unique consumer technology opportunities. You'll run out of fingers and toes before you come to a Chinese city as small as New York City, and that matters when many social products piggyback and rely on metropolitan densities as dry kindling.

The competition between tech companies in the U.S. draws scandalized chatter from the peanut gallery, but the pace at which something like Snapchat Stories was copied in the U.S. would be seen as laughably slow in China. Not only are features of competitors routinely copied within a week or two in China, employees are poached all the time in what is closer to a true approximation of a free labor market than even Silicon Valley. Knowledge moves quickly, freely.

Three things, in my observation, hold Chinese companies back from capturing more share in the international market, outside Chinese borders. Two are related, and those are an internationally appealing industrial and software design aesthetic.

It's true, people who find many Chinese software UI's to be busy and crowded can't read Chinese and thus don't understand their localized appeal to the Chinese market. As eye-tracking studies have shown (example), Chinese users scan pages differently, and why shouldn't they considering the fundamental differences between an alphabetic language like English and a pictographic one like Chinese?

Still, most of the international market can't read Chinese. In my past work with UI designers in China, I find it takes more prompting to arrive at something more broadly intuitive for, say, an American market.

The same goes for industrial design, where, akin to the denser informational aesthetic of Chinese software, a somewhat more maximalist impulse takes hold. It's still quite common to walk into an Asian electronics superstore and see display signage that lists dozens of bullet points of features in selling a product. Contrast that to the almost non-existent signage in an Apple store for the extreme opposite.

A more tangible example is something like the user interface of everyone's favorite cooking gadget, the Instant Pot. I received one as a gift last year, and no doubt, I think it's a real value at $80 or so for the base model. For how harried we all feel, a pressure cooker is way more useful a kitchen gadget than most.

However, this is the instrument panel on the front of the Instant Pot.

In practice, it's even more confusing than it appears on first glance. I won't delve into it here, but given a simple design pass, the entire UI could be made much less intimidating, much more intuitive. Given the functionality of any pressure cooker, the functionality can be reduced to a much simpler instrumentation.

These two skill gaps in software and industrial design allow for a continued Kickstarter arbitrage opportunity that slaps a more internationally appealing software and industrial design, along with the more internationally appealing marketing, on top of Shenzhen's manufacturing capabilities.

The last thing holding back more Chinese startups, in my experience, is a shortage in the professional management class. I know, I know, MBA's get a bad rap in the domestic market, but having many CEO's with engineering background at so many Chinese tech companies comes with its own drawbacks.

This management gap may be related to the style of org structure and management which others have mentioned to me as less conducive to certain types of innovation, though it's harder for me to assess without having worked inside a Chinese company.

None of this needs to matter since the Chinese market is so massive. Chinese startups can succeed wildly without ever making a peep outside their home territory. Besides, how a design aesthetic and process can seep into a country's soul remains a mystery to me, but my guess is it's about as slow-moving as trying to produce a high quality soup dumpling in a new market.

Still, I love to muse on the potential of China. In fact, there is one Chinese company that best exemplifies the potential of the country's tech market on the international stage. Last summer, a friend of mine who had worked at this company heard I was in the market for a drone and referred me to a friend who was selling an extra, lightly used Mavic Pro which he'd purchased after he thought he'd lost his original.

I don't know the first thing about flying drones, but it took me all of fifteen minutes or so to get the thing up and flying around in the air, capturing 4K video. It is an fantastic feat of engineering, probably still the single drone I'd recommend to anyone looking to get into drone photography (though I recommend getting a bundle with some extra batteries and a carrying case).

DJI had a few advantages in surging to its undisputed leadership position in the global drone market. First, this is a product category where how well the product actually works is more complex a task than in others. Many drones just don't fly that well. Being an engineering-led company is a strength here, and as long as the industrial design is optimized for flight, it doesn't really matter if your product isn't the sleekest. You won't care what it looks like when it's several hundred feet up in the air.

Second, from a software design perspective, drone UI design can piggyback on flight UI templates that have been worked out over the years. One reason I was up and running so quickly with my Mavic Pro is that the flight sticks imitate video game flight controls. The UI isn't quite as simple as I'd like, but I was fluent much more quickly than the hefty page count of the instruction manual implied.

Estimates of DJI's market share vary, but they are all well north of 50%, and most of its competitors have either left the market entirely or are struggling to stay aloft, so to speak. Here is a vertically integrated Chinese company that most definitely makes more than the cost of cheap SF lunch that Foxconn makes for each iPhone it assembles.

Now, making drones and building smartphones or writing apps are not the same skills. Drones, as exciting as they are, still aren't the type of thing I'd recommend except to photography enthusiasts. And China is a long way from dominating the consumer electronics market internationally with a massive portfolio of products from domestic, vertically integrated companies.

But the ceiling at least exists. it's not theoretical. It's more than any other country outside the U.S. can say, and how China answers it is one of the questions which will determine the relative economic power of China versus the United States in this century.

***

That China can export drones more easily than it can import, say, the software and industrial design know-how of a company like Apple is, at a higher level, a fundamental question of how we can pass along knowledge of any sort. Why are we not better at transferring know-how to industrial workers who are out of a job, why hasn't the internet produced a global leveling of industrial know-how at the country level?

Hidalgo notes:

At a finer scale, economies still lack the intimate connection between knowhow and information that is embodied in DNA and which allows biological organisms to pack knowhow so tightly. A book on engineering, art, or music can help develop an engineer, artist, or musician, but it can hardly do so with the elegance, grace, and efficiency with which a giraffe embryo unpacks its DNA to build a giraffe. The ability of economies to pack and unpack knowhow by physically embodying instructions and templates as written information is much more limited than the equivalent ability in biological organisms.
 

We are nowhere near our maximum throughput for passing on our knowledge to our fellow man, let alone across the membrane between companies and economies. In The Matrix, with a few seconds of fluttering eyelids, Keanu Reeves downloads an entire martial art into his self.

Give a man a kung-fu, you making him Neo. Teach a man to kung-fu, you make him John Wick. 

That is the dream. Asky any parent who is in the midst of trying to get a three-year-old to eat their dinner without throwing half of it on the ground and they'll nod in agreement. What is our version of nature's DNA and cell school of knowledge compression and decompression?

One of the reality TV shows which I wish existed would be one in which a variety of masters in their field compete to take absolute novices from a standing start as far as possible in a finite period of time. Instead of Top Chef, in which contestants are all successful chefs already, I want three master chefs to have to each train a handful of complete cooking dunces over a several month process, and the teacher and winning student share the pot.

Each season could have variant skills. In another, maybe the world's top three piano teachers have to train people who've never played the piano in their lives to sight-reading Chopin. Bill Belichick and Nick Saban coach two youth league football teams to see which wins a season finale scrimmage. I'm sure some of you will write me to tell me some version of a show like this exists already, but I've seen some that come close, but almost all the ones I've seen spend much too little time on the actual instruction methodology and process, and that's where all the mystery and interest lies.

In future posts, I'll delve into some of the limitations I've observed in how we pass information among people, companies, and economies, and from one generation to the next. For now, I recommend picking up Hidalgo's book, and I hope to hear from you about some of the ways you've found to help grow information around you in more efficient ways.

The Second Law of Thermodynamics

I love Steven Pinker's response to the 2017 Edge Question: what scientific term or concept ought to be more widely known?

He chose the Second Law of Thermodynamics. I'm not sure that's a concept that needs more publicity, unless you consider almost all of science underrated, which is fair. It's his reason why he chose the law which is so striking:

Why the awe for the Second Law? The Second Law defines the ultimate purpose of life, mind, and human striving: to deploy energy and information to fight back the tide of entropy and carve out refuges of beneficial order. An underappreciation of the inherent tendency toward disorder, and a failure to appreciate the precious niches of order we carve out, are a major source of human folly.
 
To start with, the Second Law implies that misfortune may be no one’s fault. The biggest breakthrough of the scientific revolution was to nullify the intuition that the universe is saturated with purpose: that everything happens for a reason. In this primitive understanding, when bad things happen—accidents, disease, famine—someone or something must have wanted them to happen. This in turn impels people to find a defendant, demon, scapegoat, or witch to punish. Galileo and Newton replaced this cosmic morality play with a clockwork universe in which events are caused by conditions in the present, not goals for the future. The Second Law deepens that discovery: Not only does the universe not care about our desires, but in the natural course of events it will appear to thwart them, because there are so many more ways for things to go wrong than to go right. Houses burn down, ships sink, battles are lost for the want of a horseshoe nail.
 
Poverty, too, needs no explanation. In a world governed by entropy and evolution, it is the default state of humankind. Matter does not just arrange itself into shelter or clothing, and living things do everything they can not to become our food. What needs to be explained is wealth. Yet most discussions of poverty consist of arguments about whom to blame for it.
 

As Shelley once wrote:

"Round the decay
Of that colossal Wreck, boundless and bare
The lone and level sands stretch far away."

Acoustic pest detection

The U.S. Grain Inspection Service, Packers, and Stockyard Admininstration’s (GISPSA) standard quality assessment method involves sieving and visually inspecting a one kilogram sample: their guidelines “consider grains infested if the representative sample contains two or more live weevils, or one live weevil and one or more other live insects injurious to stored grain, or two or more live insects injurious to stored grain.”
 
However, since the larvae of many stored product pests grow inside grain kernels, where, Fleurat-Lessard notes, their “population density may be ten times more numerous than free-living adults,” a visually-inspected“clean” sample may actually be completely infested with rice weevil larvae. To look inside grains, laboratories use X-rays or resonance spectroscopy, but these techniques are too expensive and impractical to deploy in bulk grain lots.
 
But while rice weevil larvae are invisible, they are not inaudible: the “mean sound pressure” of rice weevil larvae feeding inside a wheat kernel is 23 dB, according to the USDA Agricultural Research Service. The idea, then, is that if you could somehow design sensitive-enough acoustic probes, combined with software to match the probes’ input against a database of field recordings, you might be able to monitor insect activity in stored grain automatically and detect infestations at the larval stage.
 

I had no clue such a thing as acoustic pest detection existed. Amazing.

Building a sound library of stored food insects was equally important – the field recordings on that Insect Noise in Stored Foodstuffs CD actually form the core of current acoustic pest detection databases. Years of research have gone into classifying the characteristic sonic signatures of different pest species at different stages in their lifecycles, to the point that a computer can now compare input from a grain silo’s acoustic sensor system against a library field recordings and tell you whether the rice weevil larvae eating your wheat kernels are sixteen or eighteen days old.
 

The smartphone is a form of human augmentation, the latest version of the “bicycle for the brain” metaphor from Steve Jobs or whoever it originated with. I'm looking forward to more sensory augmentation in compact form factors in my lifetime. The ability to increase the sensitivity of my hearing and have it plug into a database of sounds for enhanced recognition would open up a whole new world. Camping would never be the same again.

Clustered regularly interspaced short palindromic repeats

For those wondering what the deal with CRISPR is, Michael Specter offers a riveting overview in the New Yorker.

The field has moved quickly. For scientists, ordering genes is almost Amazon-like in its convenience now.

Ordering the genetic parts required to tailor DNA isn’t as easy as buying a pair of shoes from Zappos, but it seems to be headed in that direction. Yan turned on the computer at his lab station and navigated to an order form for a company called Integrated DNA Technologies, which synthesizes biological parts. “It takes orders online, so if I want a particular sequence I can have it here in a day or two,” he said. That is not unusual. Researchers can now order online almost any biological component, including DNA, RNA, and the chemicals necessary to use them. One can buy the parts required to assemble a working version of the polio virus (it’s been done) or genes that, when put together properly, can make feces smell like wintergreen. In Cambridge, I.D.T. often makes same-day deliveries. Another organization, Addgene, was established, more than a decade ago, as a nonprofit repository that houses tens of thousands of ready-made sequences, including nearly every guide used to edit genes with CRISPR. When researchers at the Broad, and at many other institutions, create a new guide, they typically donate a copy to Addgene.


The field has achieved some level of efficiency with the creation of editable mice.

The vivarium at the Broad houses an entirely different kind of mouse, one that carries the protein Cas9 (which stands for CRISPR-associated nuclease) in every cell. Cas9, the part of the CRISPR system that acts like a genetic scalpel, is an enzyme. When scientists originally began editing DNA with CRISPR, they had to inject both the Cas9 enzyme and the probe required to guide it. A year ago, Randall Platt, another member of Zhang’s team, realized that it would be possible to cut the CRISPR system in two. He implanted the surgical enzyme into a mouse embryo, which made it a part of the animal’s permanent genome. Every time a cell divided, the Cas9 enzyme would go with it. In other words, he and his colleagues created a mouse that was easy to edit. Last year, they published a study explaining their methodology, and since then Platt has shared the technique with more than a thousand laboratories around the world.

The “Cas9 mouse” has become the first essential tool in the emerging CRISPR arsenal. With the enzyme that acts as molecular scissors already present in every cell, scientists no longer have to fit it onto an RNA guide. They can dispatch many probes at once and simply make mutations in the genes they want to study.


This:

He stood up and walked across the office toward his desk, then pointed at the wall and described his vision for the future of cancer treatment. “There will be an enormous chart,” he said. “Well, it will be electronic, and it will contain the therapeutic road map of every trick that cancer cells have—how they form, all the ways you can defeat them, and all the ways they can escape and defeat a treatment. And when we have that we win. Because every cancer cell starts naïve. It doesn’t know what we have waiting in the freezer for it. Infectious diseases are a different story; they share their knowledge as they spread. They learn from us as they move from person to person. But every person’s cancer starts naïve. And this is why we will beat it.”


It's a story with all the usual trappings of a technology race. Patent battles and intellectual property lawsuits. Stunning breakthroughs. And of course, the dystopia nightmares that seem to accompany genetics more than any other form of science.

Doudna is a highly regarded biochemist, but she told me that not long ago she considered attending medical school or perhaps going into business. She said that she wanted to have an effect on the world and had begun to fear that the impact of her laboratory research might be limited. The promise of her work on CRISPR, however, has persuaded her to remain in the lab. She told me that she was constantly amazed by its potential, but when I asked if she had ever wondered whether the powerful new tool might do more harm than good she looked uncomfortable. “I lie in bed almost every night and ask myself that question,” she said. “When I’m ninety, will I look back and be glad about what we have accomplished with this technology? Or will I wish I’d never discovered how it works?”

Her eyes narrowed, and she lowered her voice almost to a whisper. “I have never said this in public, but it will show you where my psyche is,” she said. “I had a dream recently, and in my dream”—she mentioned the name of a leading scientific researcher—“had come to see me and said, ‘I have somebody very powerful with me who I want you to meet, and I want you to explain to him how this technology functions.’ So I said, Sure, who is it? It was Adolf Hitler. I was really horrified, but I went into a room and there was Hitler. He had a pig face and I could only see him from behind and he was taking notes and he said, ‘I want to understand the uses and implications of this amazing technology.’ I woke up in a cold sweat. And that dream has haunted me from that day. Because suppose somebody like Hitler had access to this—we can only imagine the kind of horrible uses he could put it to.”

The cold hard facts

When your Jeep spins lazily off the mountain road and slams backward into a snowbank, you don't worry immediately about the cold. Your first thought is that you've just dented your bumper. Your second is that you've failed to bring a shovel. Your third is that you'll be late for dinner. Friends are expecting you at their cabin around eight for a moonlight ski, a late dinner, a sauna. Nothing can keep you from that.
 
Driving out of town, defroster roaring, you barely noted the bank thermometer on the town square: minus 27 degrees at 6:36. The radio weather report warned of a deep mass of arctic air settling over the region. The man who took your money at the Conoco station shook his head at the register and said he wouldn't be going anywhere tonight if he were you. You smiled. A little chill never hurt anybody with enough fleece and a good four-wheel-drive.
 

Thus begins a chilling piece (at the exact moment I wrote that, the pun was not intended, but who will believe me?) on what it's like to freeze to death. This was written in 2004, but so much of the web is evergreen, hard as it is to hear above the ever cresting feed.

I thought of this when I went out for my first morning bike ride of the Bay Area fall/winter. This was an unusually warm summer, but the winter chill seemed to come overnight. I walked out the door of my apartment into the morning air and had to suppress the urge to cry. I turned right back around and went back in to don a snowsuit. Over time, my body will acclimate, but for now, it feels as if I'm wading out into a frozen tundra.

At 85 degrees, those freezing to death, in a strange, anguished paroxysm, often rip off their clothes. This phenomenon, known as paradoxical undressing, is common enough that urban hypothermia victims are sometimes initially diagnosed as victims of sexual assault. Though researchers are uncertain of the cause, the most logical explanation is that shortly before loss of consciousness, the constricted blood vessels near the body's surface suddenly dilate and produce a sensation of extreme heat against the skin.
 
All you know is that you're burning. You claw off your shell and pile sweater and fling them away.
 
But then, in a final moment of clarity, you realize there's no stove, no cabin, no friends. You're lying alone in the bitter cold, naked from the waist up. You grasp your terrible misunderstanding, a whole series of misunderstandings, like a dream ratcheting into wrongness. You've shed your clothes, your car, your oil-heated house in town. Without this ingenious technology you're simply a delicate, tropical organism whose range is restricted to a narrow sunlit band that girds the earth at the equator.
 
And you've now ventured way beyond it.

Science is hard

Taken together, headlines like these might suggest that science is a shady enterprise that spits out a bunch of dressed-up nonsense. But I’ve spent months investigating the problems hounding science, and I’ve learned that the headline-grabbing cases of misconduct and fraud are mere distractions. The state of our science is strong, but it’s plagued by a universal problem: Science is hard — really fucking hard.
 
If we’re going to rely on science as a means for reaching the truth — and it’s still the best tool we have — it’s important that we understand and respect just how difficult it is to get a rigorous result. I could pontificate about all the reasons why science is arduous, but instead I’m going to let you experience one of them for yourself. Welcome to the wild world of p-hacking.
 

A very important piece at 538.com on p-values and the likely prevalence of p-hacking.

The p-value reveals almost nothing about the strength of the evidence, yet a p-value of 0.05 has become the ticket to get into many journals. “The dominant method used [to evaluate evidence] is the p-value,” said Michael Evans, a statistician at the University of Toronto, “and the p-value is well known not to work very well.”
 
...
 
But that doesn’t mean researchers are a bunch of hucksters, a la LaCour. What it means is that they’re human. P-hacking and similar types of manipulations often arise from human biases. “You can do it in unconscious ways — I’ve done it in unconscious ways,” Simonsohn said. “You really believe your hypothesis and you get the data and there’s ambiguity about how to analyze it.” When the first analysis you try doesn’t spit out the result you want, you keep trying until you find one that does. (And if that doesn’t work, you can always fall back on HARKing — hypothesizing after the results are known.)
 

The larger lessons apply not just to science. Journalism is hard, especially investigative journalism. You can spend months reporting a piece only to find no real striking narrative, no clear conclusions of note. And yet, if you have to fill a certain number of pages every day...

In tech, really successful and/or counterintuitive A/B test results are passed around like koans. However, anyone who has done enough A/B testing in the tech world knows that most experiments show no statistically significant results. To design a test that won't show the obvious and that will reveal some hidden truth is not easy.

All data suggests most of us should hold our unproven beliefs more loosely than we're inclined to. Who first came up with the saying “Strong opinions, weakly held” (sometimes “loosely” is substituted). Most of us are good at the first half, not so good at the second, a dangerous combination when it turns out that truth is low yield.

Some things might help. One is something of a reference that is a collection of links to all studies that have tried to answer a particular question along with a summary of the current state of thinking. For example, does drinking a glass of red wine a day improve your health? Why are Americans obese? Does eating a multivitamin every day really do anything for your health? What's the best exercise to improve core strength? And so on. Imagine something like the genetic offspring of Vox, Wikipedia, and Richard Feynman.

Another is something like Github but for research data from all these studies. 538's small experiment widget in this piece was a simplified example of the type of tool that might enable more people to get experience and a deeper understanding of the craft of designing studies and the slippery nature of truth. Also, the more people that can analyze a data set, the greater the likelihood that biases of different types balance each other out and that mistakes are caught. Strong hypotheses can often lead one to control for the very variable that explains a result.

The web is so sprawling, information so infinite now, we need more structured ways to traverse it intelligibly. It's no coincidence one of the words that's entered our vocabulary this past year is “explainer” (here is an explainer on the term explainer). We have so much flow, we need more stock.

Game theory of life

In what appears to be the first study of its kind, computer scientists report that an algorithm discovered more than 50 years ago in game theory and now widely used in machine learning is mathematically identical to the equations used to describe the distribution of genes within a population of organisms. Researchers may be able to use the algorithm, which is surprisingly simple and powerful, to better understand how natural selection works and how populations maintain their genetic diversity.

By viewing evolution as a repeated game, in which individual players, in this case genes, try to find a strategy that creates the fittest population, researchers found that evolution values both diversity and fitness.

Some biologists say that the findings are too new and theoretical to be of use; researchers don’t yet know how to test the ideas in living organisms. Others say the surprising connection, published Monday in the advance online version of the Proceedings of the National Academy of Sciences, may help scientists understand a puzzling feature of natural selection: The fittest organisms don’t always wipe out their weaker competition. Indeed, as evidenced by the menagerie of life on Earth, genetic diversity reigns.

Fascinating. It's tempting to try to imagine where the value of both fitness and diversity might extend outside of genetics. Clearly it has value in finance in portfolio theory; perhaps it matters in organizations, too? Personal ideology? Friend selection? Team construction?

My dad is smart

Through a Google search, my dad found a physics research paper he wrote years ago. The title: “Nuclear magnetic resonance study of the diffusion of bound and free fluorine interstitials in alkaline-earth fluorides doped with trivalent impurities.”

The abstract:

When alkaline-earth fluorides are doped with trivalent impurities, interstitial fluorines are created to maintain charge neutrality. We have performed NMR dipolar-energy relaxation measurements over a wide temperature range and have observed the diffusion of both free and locally bound fluorine interstitials F−(i) in the extrinsic region. We have observed that these motions depend strongly on the host-crystal lattice parameter. In particular, we have observed that the motion of F−(i) which is at the nearest neighbor (nn) site to the trivalent impurity dominates the relaxation above room temperature in CaF2: Y3+ but is unobserved in BaF2: Y3+. In addition, a second type of bound F−(i) motion, characterized by a much smaller activation energy, appears over a very narrow temperature range (150-185°C) in CaF2: Y3+, but over a large temperature range (below room temperature to 130°C) in BaF2: Y3+. SrF2: Y3+ shows similar behavior over a temperature range (54-160°C), which is intermediate between that of CaF2: Y3+ and BaF2: Y3+. Possible explanations in terms of the motion of a more remotely bound F−(i) [e.g., at a next-nearest-neighbor (nnn) site] and the motion of F−(i) near clusters of dipoles are discussed. We measured activation energies for all these F−(i)motions. A comparison of our results with those by other techniques (specifically, EPR, ENDOR, optical spectroscopy, ionic conductivity, ionic thermocurrent, dielectric and anelastic loss) is also given.

It's a strange feeling realizing your dad will always be smarter than you by a wide margin. I still have to teach him how to use his new Mac, though.