WTF?

WTF? Plot Summary

Introduction

In the late 1990s, a small group of software developers gathered in Palo Alto to coin a new term: "open source." This seemingly minor linguistic shift from "free software" would spark one of the most profound power transformations in economic history. What began as a movement to make code accessible evolved into a fundamental restructuring of how value is created and distributed in our digital society. From Linux to the App Store, from Uber to algorithmic trading, we've witnessed a remarkable journey where the rules of economic power have been continuously rewritten. This historical journey reveals how digital technologies have created entirely new governance structures that often operate outside traditional regulatory frameworks. As platforms like Amazon and Facebook grew from simple websites to market-defining ecosystems, they developed their own rule systems that now govern significant portions of our economic lives. Understanding this evolution helps us recognize that today's algorithmic systems aren't merely technical tools but power structures that encode specific values and priorities. For policymakers, business leaders, and citizens seeking to navigate our increasingly digital future, this exploration offers crucial insights into how we might harness technological innovation while ensuring its benefits are widely shared.

Chapter 1: Open Source Foundations: Collaborative Innovation (1998-2005)

The period from 1998 to 2005 marked a fundamental shift in how software was created and distributed, establishing patterns that would later transform the entire economy. The term "open source" itself was coined in 1998 at a summit organized by Tim O'Reilly, deliberately replacing the more ideologically charged "free software" terminology. This seemingly simple rebranding represented a profound shift in thinking about how value could be created through collaboration rather than competition. At the heart of this transformation was a radical new development model. As Eric Raymond articulated in his influential essay "The Cathedral and the Bazaar," open source projects like Linux demonstrated that software could be developed through a decentralized "bazaar" model rather than the traditional "cathedral" approach of proprietary software companies. Raymond's key insight that "given enough eyeballs, all bugs are shallow" captured how distributed collaboration could solve complex problems more effectively than centralized control. This period saw Linux evolve from a hobbyist project to enterprise-grade software, while Apache became the world's most widely used web server. The economic implications were revolutionary. Open source demonstrated that valuable digital goods could be created outside traditional market mechanisms. Companies like Red Hat pioneered business models that created value not by selling proprietary software but by providing services around freely available code. Meanwhile, internet infrastructure increasingly relied on open protocols and standards that weren't owned by any single entity. By 2005, much of the internet ran on open source technologies like Apache, MySQL, PHP, and Linux – creating billions in economic value while remaining freely accessible. This era also witnessed the birth of Web 2.0, a term popularized by O'Reilly that captured the evolution of the internet from static websites to interactive platforms. The principles of Web 2.0 – the internet as a platform, harnessing collective intelligence, and data as a competitive advantage – would become foundational to the platform businesses that followed. Early social media platforms, blogging tools, and collaborative sites like Wikipedia demonstrated that users could be both consumers and creators, fundamentally changing the relationship between producers and audiences. The collaborative innovation model pioneered during this period established patterns that would later transform industries far beyond software. The principles of distributed contribution, permissionless innovation, and shared infrastructure created new possibilities for organizing economic activity. As Tim O'Reilly noted, "The architecture of the system determines what's possible." By creating architectures that enabled broad participation and shared value creation, the open source movement laid the groundwork for the platform economy that would follow, establishing both its tremendous creative potential and its challenging governance questions.

Chapter 2: Platform Revolution: The Rise of Digital Marketplaces

Between 2005 and 2012, a new economic model emerged that would fundamentally transform how businesses create and capture value. This period saw the rise of digital platforms – multi-sided marketplaces that connect producers and consumers in ways that create exponentially more value than traditional linear businesses. Companies like Amazon, Apple's App Store, and Facebook demonstrated that the most valuable businesses weren't necessarily those that created products, but those that facilitated connections. At the heart of this transformation was a shift in organizational structure. Traditional businesses operated as pipelines, with clear value chains flowing from suppliers to the company to customers. Platform businesses, by contrast, operated as networks, creating value by facilitating exchanges between external producers and consumers. When Apple launched the App Store in 2008, it wasn't just selling its own software but creating a marketplace where thousands of developers could reach millions of users. This approach allowed platforms to scale at unprecedented rates – by 2012, the App Store had facilitated over 25 billion app downloads. The economics of these platforms proved revolutionary. While traditional businesses faced diminishing returns to scale as they grew, platforms experienced the opposite – network effects that made them more valuable as more people used them. Facebook became more valuable to users as more of their friends joined. Marketplaces like eBay became more attractive to buyers as more sellers participated, and vice versa. This dynamic created winner-take-all markets where the leading platform could achieve dominant market share, explaining why by 2012, Google controlled over 90% of search in many markets. These platforms also demonstrated a remarkable ability to scale without corresponding increases in marginal costs. As Jeff Bezos explained in his famous 2002 "platform mandate," Amazon rebuilt its infrastructure as a set of well-documented services that could be accessed through standardized APIs. This approach not only made Amazon's internal operations more efficient but eventually led to Amazon Web Services (AWS), transforming how businesses accessed computing resources. By 2012, AWS was hosting Netflix, Instagram, and thousands of other companies, demonstrating how platforms could create entirely new business categories. The platform model extended beyond pure digital services into the physical world. By 2012, Airbnb had listed over 200,000 properties worldwide without owning a single hotel room. This approach of replacing ownership with access, and using algorithms to manage resources more efficiently than traditional coordination methods, would soon transform industries from transportation to food delivery. The platforms weren't just digitizing existing business models – they were fundamentally restructuring economic relationships. By 2012, these platforms had begun to reshape the economy and society in profound ways. They created new forms of work, disrupted established industries, and raised important questions about market power and regulation. The success of these platforms wasn't just technological but organizational – they had discovered a fundamentally different way of creating and capturing value in the digital age, one that would continue to expand its reach in the years that followed.

Chapter 3: On-Demand Economy: Transforming Service Industries

Between 2009 and 2016, a new economic model emerged that extended digital platforms into the physical world, fundamentally transforming how services are delivered. At the forefront of this revolution was Uber, founded in 2009 by Travis Kalanick and Garrett Camp with a simple premise: allowing users to summon a black car with the tap of a smartphone. What began as a luxury service quickly evolved into a paradigm shift that would reshape service industries worldwide. The on-demand model introduced several revolutionary elements. First was the magical user experience – the ability to summon a service precisely when and where needed, with transparent pricing and frictionless payment. Second was the algorithmic management of a distributed workforce, using surge pricing to balance supply and demand in real-time. Third was the two-sided marketplace that created network effects – more drivers meant shorter wait times for riders, which attracted more riders, which in turn attracted more drivers. By 2014, Uber was operating in over 200 cities and valued at $18 billion. The model quickly spread beyond transportation. By 2015, DoorDash and Instacart were applying similar principles to food and grocery delivery. TaskRabbit created a marketplace for small jobs and errands. Each of these services replaced traditional scheduling and dispatch with algorithmic matching and real-time coordination. The smartphone, combining GPS, payments, and always-on connectivity, became the remote control for the physical world. As venture capitalist Shervin Pishevar noted in 2012, "Uber is not a transportation company, it's a technology platform that can transform any service industry." The impact on labor markets was profound. By 2016, an estimated 4% of American adults had earned money through on-demand platforms. These platforms created flexible earning opportunities, allowing people to work when and where they chose. However, they also shifted risks from companies to workers by classifying them as independent contractors rather than employees. This arrangement eliminated traditional benefits like health insurance and minimum wage protections. The model challenged existing regulatory frameworks designed for an era of stable employment relationships. For consumers, the on-demand economy dramatically reduced transaction costs – the friction involved in finding, negotiating with, and paying service providers. Activities that once required advance planning became available on demand. This convenience came with trade-offs, however. The platforms collected vast amounts of data about both service providers and customers, raising privacy concerns. Their algorithmic management systems sometimes produced unintended consequences, from surge pricing during emergencies to racial discrimination in ride acceptance. By 2016, the on-demand economy had become a significant economic force, with Uber alone completing over 2 billion rides worldwide. The model had demonstrated that combining smartphones, payment systems, GPS, and algorithmic management could transform almost any service industry. The question was no longer whether on-demand services would reshape the economy, but how far the transformation would extend and what regulatory frameworks would evolve to govern these new business models. The gig economy had arrived, bringing both unprecedented convenience and challenging questions about the future of work.

Chapter 4: Algorithmic Decision-Making: The New Power Structure

Between 2010 and 2018, algorithms evolved from behind-the-scenes technical tools to powerful decision-making systems that increasingly shaped our economy, society, and daily lives. This shift represented a profound transfer of authority from human judgment to computational processes, creating what some scholars have called "algocracy" – rule by algorithm. Understanding this transformation requires looking beyond the technical aspects of algorithms to their role as governance structures. At the most visible level, algorithms began determining what information we see online. By 2016, Facebook's News Feed algorithm was selecting content for over 1.7 billion monthly users, while Google's search algorithms processed over 3.5 billion searches daily. These algorithms weren't neutral technical tools but powerful gatekeepers, shaping public discourse and information access. Their design choices – what signals to prioritize, what content to promote or demote – became de facto policy decisions with far-reaching implications. As Facebook engineer Ansha Jain acknowledged in 2016, "We have a responsibility to make sure Time Well Spent is the goal of everything we build." The fitness functions of these algorithms – what they were designed to optimize – revealed their underlying values and priorities. Facebook optimized for "engagement," measured by clicks, likes, shares, and comments. This seemingly technical choice had profound consequences, as content that triggered strong emotional reactions (particularly outrage) spread more widely than nuanced discussion. By 2016, studies were showing that false news stories often outperformed accurate ones on social media precisely because they were optimized for engagement rather than accuracy. Beyond information, algorithms increasingly made consequential decisions about people's lives. By 2017, over 70% of resumes were screened by automated systems before human eyes ever saw them. Credit scoring algorithms determined who could get loans and at what interest rates. Criminal justice algorithms assessed recidivism risk and influenced sentencing and parole decisions. Healthcare algorithms prioritized patients for treatments. In each domain, complex mathematical models made judgments that were often opaque even to their creators. These systems raised profound questions about accountability and fairness. Machine learning algorithms, trained on historical data, often reproduced and amplified existing biases. A 2016 ProPublica investigation found that a widely used criminal risk assessment tool was nearly twice as likely to falsely flag Black defendants as high risk compared to white defendants. As Cathy O'Neil argued in her 2016 book "Weapons of Math Destruction," these systems could create pernicious feedback loops that reinforced inequality while claiming the mantle of objective science. The governance implications were equally significant. Traditional regulatory approaches struggled to address algorithmic systems that evolved constantly and operated at massive scale. The companies deploying these algorithms often claimed they couldn't fully explain how their systems made decisions, creating a "black box" problem for oversight. Meanwhile, the concentration of algorithmic power in a handful of technology companies raised concerns about market dominance and democratic accountability. By 2018, it had become clear that algorithms weren't just technical tools but governance systems that encoded values, made policy choices, and exercised power. The question wasn't whether algorithms would make important decisions, but who would control them, what values they would embody, and how society could ensure they served the public good rather than narrow commercial interests. As legal scholar Frank Pasquale wrote, "The era of algorithmic accountability had begun."

Chapter 5: Human-Machine Collaboration: Redefining Work

Between 2015 and 2020, the relationship between humans and machines in the workplace underwent a profound transformation. As artificial intelligence advanced rapidly, two competing narratives emerged: one emphasizing job displacement through automation, the other focusing on human-machine collaboration. This tension between replacement and enhancement became central to understanding technology's impact on labor markets and designing appropriate policy responses. The automation narrative gained prominence as machine learning systems demonstrated increasingly sophisticated capabilities. In 2016, AlphaGo defeated the world champion at Go, a game long considered too complex for machines to master. Self-driving vehicles began testing on public roads. Image recognition systems outperformed humans on specific tasks. These advances led to alarming predictions about job losses, with a widely-cited 2013 Oxford study suggesting that up to 47% of US jobs were at risk of automation within two decades. However, a more nuanced view emerged from examining how technologies were actually being deployed in workplaces. In many cases, the most effective implementations paired human judgment with machine capabilities rather than replacing humans entirely. By 2018, radiologists were working with AI systems that flagged potential abnormalities but relied on human expertise for final diagnosis. Customer service representatives used chatbots to handle routine inquiries while focusing their attention on complex problems requiring empathy and creativity. As MIT professor Erik Brynjolfsson noted, "The most productive systems are neither human-only nor machine-only, but human-machine partnerships." This augmentation approach recognized that humans and machines excel at different types of tasks. Machines process vast amounts of data and execute well-defined procedures with speed and consistency. Humans excel at contextual understanding, ethical reasoning, creativity, and interpersonal communication. By 2019, companies like Stitch Fix were demonstrating the power of this complementary approach, with human stylists working alongside recommendation algorithms to provide personalized clothing selections at scale. The distinction between automation and augmentation had profound implications for skills development and education. If the goal was replacement, then preparing workers meant training them for jobs least susceptible to automation. If the goal was augmentation, then education needed to focus on uniquely human capabilities that complement machine intelligence – critical thinking, creativity, emotional intelligence, and adaptability. By 2020, leading companies were investing heavily in reskilling programs that helped employees develop these complementary capabilities. Historical patterns offered some reassurance. Previous waves of technological change had eliminated specific jobs but created new ones, often in greater numbers. The introduction of ATMs reduced the number of tellers per bank branch but led to an increase in the total number of bank employees as branches expanded and tellers took on more complex customer service roles. Similarly, when law firms adopted document search technology, they employed more lawyers who could focus on higher-value tasks rather than document review. By 2020, the most promising vision of the future of work centered on human-machine partnerships that enhanced productivity while creating meaningful work. This required intentional design choices that put humans at the center, using technology to remove drudgery while expanding human capabilities. As Microsoft CEO Satya Nadella put it in 2019, "The question isn't just what computers can do, it's what computers should do." The challenge for society was not just technological but institutional – creating education systems, labor market policies, and social safety nets that supported workers through this transition.

Chapter 6: Government as Platform: Reimagining Public Services

Between 2009 and 2018, a new vision of government emerged that applied platform thinking to public services. This approach, often called "Government as a Platform," reimagined government not as a provider of monolithic services but as an enabler of innovation and value creation by citizens, businesses, and civil society. This represented a fundamental shift from the traditional "vending machine" model of government, where citizens pay taxes and receive standardized services in return. The concept gained momentum following the 2008 election of Barack Obama, often described as the first "Internet president" due to his campaign's effective use of digital tools. In 2009, the US government launched Data.gov, making thousands of government datasets freely available in machine-readable formats. This initiative built on earlier efforts like the digitization of SEC filings, which had demonstrated how government data could create economic value when made accessible to innovators. By 2015, over 200,000 datasets were available, powering applications from climate research to consumer protection tools. The UK government went further with the creation of the Government Digital Service (GDS) in 2011. Under the leadership of Mike Bracken, the GDS completely redesigned the UK government's web presence around user needs rather than departmental structures. Their design principles emphasized starting with user needs, doing less (focusing on what only government can do), and making things open. By 2015, this approach had not only improved citizen experience but saved the UK government over £1.7 billion through more efficient digital services. In the United States, the healthcare.gov crisis of 2013 – when the Obama administration's signature healthcare marketplace website failed spectacularly at launch – became a catalyst for digital government reform. The rescue effort, led by a team of Silicon Valley engineers, demonstrated the value of modern software development practices in government. This led to the creation of the United States Digital Service and 18F, government agencies dedicated to improving digital services using user-centered design, agile development, and modern technology. By 2016, these organizations had improved dozens of critical government services, from veterans' benefits to immigration applications. At the local level, Code for America, founded by Jennifer Pahlka in 2009, applied similar principles to city governments. The organization placed technologists in city halls across America, building simple, effective digital services that demonstrated how technology could make government more responsive and efficient. Their work on streamlining food stamp applications in California showed how digital approaches could not only improve service delivery but inform better policy making. As Pahlka noted, "When government services are delivered digitally by default, it's not just more efficient, it actually changes what's possible." The platform approach to government extended beyond technology to governance itself. Open data initiatives enabled citizens to hold government accountable and build services government hadn't imagined. Participatory budgeting allowed communities to allocate portions of public budgets. By 2018, cities like Barcelona were pioneering "digital democracy" platforms that enabled citizens to propose and develop policy initiatives that the government committed to implementing if they received sufficient support. By 2018, the vision of government as platform had demonstrated significant promise, though implementation remained uneven. The approach required not just new technology but cultural change within government institutions accustomed to controlling rather than enabling. It also raised important questions about privacy, security, and digital inclusion. Nevertheless, the platform model offered a path to more responsive, efficient, and participatory government at a time when trust in public institutions was declining in many democracies.

Chapter 7: Value Creation vs. Extraction: Designing Sustainable Systems

Between 2016 and 2020, a fundamental question emerged about the digital economy: was it primarily creating new value or merely extracting and concentrating existing value? This distinction between value creation and value extraction became central to debates about the sustainability of our economic system and how to ensure technological progress benefits society broadly rather than narrowly. The platform economy demonstrated tremendous capacity for value creation. By 2020, Amazon's marketplace supported over 1.7 million small and medium-sized businesses worldwide. Apple's App Store had enabled developers to earn over $155 billion since its launch. These platforms created new possibilities for entrepreneurs to reach global markets with minimal capital investment. As venture capitalist Marc Andreessen noted in 2011, "Software is eating the world" – creating new industries while transforming existing ones. However, concerns about value extraction grew alongside these successes. By 2020, the five largest tech companies represented over 20% of the S&P 500's total value, while employing a fraction of the workforce of earlier industrial giants. Financial markets increasingly rewarded companies for extracting value through financial engineering rather than creating it through innovation and investment. According to William Lazonick's research, between 2009 and 2018, S&P 500 companies spent over $4.3 trillion on stock buybacks – often exceeding their investments in research and development. The distinction between value creation and extraction became particularly clear in labor markets. While digital platforms created flexible work opportunities, they often classified workers as independent contractors rather than employees, shifting risks from companies to individuals while avoiding traditional employer obligations. By 2020, an estimated 36% of American workers participated in the gig economy, but many struggled with income volatility and lack of benefits. As labor economist David Weil observed, this represented a "fissuring" of the workplace that separated the economic benefits of work from the responsibilities toward workers. Data emerged as another contested domain between creation and extraction. By collecting vast amounts of user data, platforms gained valuable insights that improved their services but also created information asymmetries that could be exploited. As economist Glen Weyl noted, users who contributed data received little direct compensation, while the platforms capturing that data achieved unprecedented market valuations. This raised questions about whether current arrangements fairly distributed the value created through data. Alternative models demonstrated that technology could enable more balanced value distribution. Platform cooperatives like Stocksy (a stock photography platform owned by its contributors) showed how digital platforms could distribute ownership and governance among participants rather than concentrating them with investors. Open-source projects like Linux continued to create tremendous value while making their outputs freely available rather than extracting rents. Companies like Patagonia demonstrated that businesses could succeed financially while prioritizing environmental sustainability and worker well-being. By 2020, the challenge was clear: how to design economic systems that encouraged value creation while limiting value extraction. This required rethinking multiple aspects of our economic architecture – from corporate governance and labor law to data rights and competition policy. As economist Mariana Mazzucato argued, we needed to move beyond simplistic distinctions between government and markets to focus on how different institutional arrangements shaped whether value was created or extracted. The most promising approaches recognized that sustainable prosperity required aligning the incentives of businesses, workers, communities, and the environment rather than privileging any single stakeholder.

Summary

The evolution from open source software to algorithmic governance represents one of the most profound power shifts in economic history. Throughout this journey, we've witnessed how collaborative innovation models pioneered in software development expanded to reshape entire industries through platforms, on-demand services, and algorithmic decision-making. What began as technical innovations quickly became new governance structures that now determine who creates value, how it's distributed, and who benefits from technological progress. The central tension running through this transformation involves balancing innovation with inclusion, efficiency with equity, and automation with augmentation. The most successful approaches don't simply optimize for short-term financial returns but create sustainable value for multiple stakeholders. As we navigate an increasingly algorithmic future, our most important task isn't predicting precisely how technology will evolve, but ensuring that our economic institutions direct technological development toward broadly shared prosperity rather than concentrated power. This requires reimagining not just business models but governance structures, educational systems, and social contracts to ensure that digital systems augment human potential rather than diminishing it. By consciously designing these systems to create more value than they capture, we can harness the tremendous potential of digital technologies while avoiding their most dangerous pitfalls.

About Author

Loading...

Tim O'Reilly

Read more