Chapter 4
MY DEFINITION OF THE METAVERSE SHOULD PROVIDE some insight into why it is often thought of, and fairly described as, a successor to the mobile internet. The Metaverse will require the development of new standards and creation of new infrastructure, potentially require overhauls to the long-standing Internet Protocol Suite, involve the adoption of novel devices and hardware, and might even alter the balance of power between technology giants, independent developers, and end users.
The enormity of this transformation also explains why companies are repositioning themselves in expectation of the Metaverse even though its arrival remains far off and its effects largely unclear. As shrewd business leaders know well, every time a new computing and networking platform emerges, the world and the companies that lead it are forever changed.
In the mainframe era, which ran from the 1950s through the 1970s, the dominant computing operating systems were those of “IBM and the Seven Dwarfs,” typically defined as Burroughs, Univac, NCR, RCA, Control Data, Honeywell, General Electric. The personal computer era, which began in earnest in the 1980s, was briefly led by IBM and its operating system. However, the eventual winners were new entrants, most notably Microsoft, whose Windows operating system and Office software suite ran on nearly every PC in the world, and manufacturers such as Dell, Compaq, and Acer. In 2004, IBM exited the business altogether, selling its ThinkPad line to Lenovo. The story of the mobile era takes a similar shape. New platforms rose or emerged, namely those of Apple’s iOS and Google’s Android, with Windows falling out of the category altogether and PC-era manufacturers displaced by new entrants such as Xiaomi and Huawei.*
Indeed, generational changes in computing and networking platforms routinely disrupt even the most stagnant and protected categories. In the 1990s, for example, chat services such as AOL Instant Messenger and ICQ quickly established text-based communications platforms which rivaled the customer bases and usage of many telephone companies and even postal services. In the 2000s, these services were surpassed by those focused on live audio, such as Skype, which also connected to traditional and offline phone systems. The mobile era saw a new crop of leaders such as WhatsApp, Snapchat, and Slack. These players didn’t just focus on offering Skype but were made for mobile devices. They built services predicated upon different usage behaviors, needs, and even communication styles.
WhatsApp, for example, is intended for nearly constant use—not scheduled or occasional calls, as was the case with Skype—and it is a forum where emojis articulate more than typed words. Whereas Skype was originally built around the ability to make low-to-no-cost calls to the traditional “Public Switched Telephone Network” (i.e., telephones connected to telephone lines), WhatsApp skipped this feature altogether. Snapchat saw mobile communication as being image-first and the front-facing camera on smartphones as more important than the more frequently used (and higher-resolution) back camera, and built numerous AR lenses to enhance that experience as a result. Slack, for its part, built a productivity-based tool for business with programmatic integration into various productivity tools, online services, and more.
Another example comes from the even more regulated and stagnant payments space. In the late 1990s, peer-to-peer digital payment networks such as Confinity and Elon Musk’s X.com, which merged to form PayPal, rapidly became consumers’ preferred method of sending money. By 2010, PayPal was processing nearly $100 billion in payments per year. A decade later, this sum exceeded $1 trillion (in part due to the acquisition of Venmo in 2012).
We can already see precursors to the Metaverse. In platforms and operating systems, the most talked about contenders are virtual world platforms like Roblox and Minecraft, and real-time rendering engines such as Epic Games’ Unreal engine and Unity Technologies’ eponymous engine. These all run on an underlying operating system, such as iOS or Windows, but they often intermediate these platforms from developers and end-users. Discord, meanwhile, operates the largest communications platform and social network focused on video gaming and virtual worlds. In 2021 alone, over $16 trillion was settled through blockchain/cryptocurrency networks, which to many experts are foundational enablers of the Metaverse (more on this in Chapter 11). Visa, as a point of contrast, processed an estimated $10.5 trillion.1
Understanding the Metaverse as the “next-generation internet” helps explain much more than its potential for disruption. Consider, once again, that there is no plural form of the term “internet.” There is no “Facebook internet” or “Google internet.” Instead, Facebook and Google operate platforms, services, and hardware that in turn operate on the internet—a literally defined “network of networks”† operating independently, with different technical stacks, but sharing common standards and protocols. There were no strict technical obstacles to a single company developing, then owning and controlling, the Internet Protocol Suite (and some, such as IBM, tried to push their own proprietary suite as part of the so-called Protocol Wars). However, most generally believe this would have led to a smaller, less lucrative, less innovative internet.‡ We should expect the establishment of the Metaverse to be broadly similar to that of the internet. Many will try to build or co-opt the Metaverse. One of these groups might even succeed, as Sweeney fears. However, it’s more likely that the Metaverse will be produced through the partial integration of many competing virtual world platforms and technologies. This process will take time. It will also be imperfect, inexhaustible, and face significant technical limitations as a result. But it is the future we should hope for and work towards.
Moreover, the Metaverse will not replace or fundamentally alter the internet’s underlying architecture or protocol suite. Instead, it will evolve to build on top of it in a way that will feel distinctive. Think about the “current state” of the internet. We refer to it as the mobile internet era, yet most internet traffic is still transmitted via fixed-line cables—even for data sent from and to mobile devices—and mostly runs on standards, protocols, and formats designed decades ago (though they’ve evolved since). We also continue to use some software and hardware designed for the early internet—such as Windows or Microsoft Office—that have evolved since, but are broadly unchanged from decades ago. Despite this, it’s clear that the “mobile internet era” is distinct from the predominantly fixed-line internet era of the 1990s and early 2000s. We now primarily use different devices (made by different companies) in new places, for different purposes, using different types of software (mostly apps, rather than general purpose software and web browsers).
We also recognize that the internet is a bundle of many different “things.” To interact with the internet, the average person typically uses a web browser or app (software), which they access through a device that can itself connect to “the internet” using various chipsets, all of which communicate using various standards and common protocols, which are transmitted through physical networks. Each of these areas collectively enable internet experiences. No one company could drive end-to-end improvements in the internet—even if they operated the entire Internet Protocol Suite.
Why Video Games Are Driving the Next Internet
If the Metaverse is indeed a successor to the internet, it might seem odd that its pioneers come from the video gaming industry. After all, the arc of the internet thus far is quite different.
The internet originated in government research labs and universities. Later, it expanded into enterprise, then small-to-medium businesses, and later still, consumers. The entertainment industry was arguably one of the last segments of the global economy to embrace the internet, with the “Streaming Wars” only really beginning in 2019—nearly 25 years after the first public demonstration of streaming video. Even audio, one of the simplest media categories to deliver over IP, remains a mostly non-digital medium, with terrestrial radio, satellite radio, and physical media comprising nearly two-thirds of US recorded music revenues in 2021.
The mobile internet was not led by government, but its arc was broadly the same. When it launched in the early 1990s, usage and software development was concentrated in government and enterprise, then by the late 1990s and early 2000s, small-to-medium businesses. Only after 2008, with the launch of the iPhone 3G, did the mass market adopt it, with consumer-centric apps emerging, for the most part, in the decade that followed.
If we look more closely at this history, we can see why gaming, a $180-billion leisure industry, seems poised to alter the $95-trillion world economy. The key is to consider the role of constraints in all technical development.
When the internet emerged, bandwidth was limited, latency was considerable, and computer memory and processing power scarce. This meant that only small files could be sent and it still took a great deal of time. Almost all consumer use cases, such as photo sharing, video streaming, and rich communications were impossible. But the primary business need—sending messages and basic files (an unformatted Excel sheet, stock purchase orders)—was exactly what the internet was designed to support. The immensity of the service economy, and the importance of managerial functions in the goods economy, was such that even modest productivity enhancements were extraordinarily valuable. Mobile was similar. Early devices couldn’t play games or send photos—and streaming a video or FaceTime call was out of the question. However, push email was orders of magnitude more helpful than pager notifications or live phone calls.
Given their complexity, it should be obvious that real-time rendered 3D virtual worlds and simulations were even more constrained by the early decades of the personal computer and internet than almost all other types of software and programs. To this end, governments, enterprises, and small-to-medium businesses had little to no use for graphics-based simulations. A virtual world that can’t realistically simulate a fire isn’t helpful for firefighters, a bullet that doesn’t curve with gravity doesn’t aid military snipers, and an architectural firm can’t design a building based on the generic idea of “heat from the sun.” But video games—games—don’t need realistic fire, gravity, or thermodynamics. What they need is to be fun. And even an 8-bit, monochromic game can be fun. The consequence of this fact has compounded for nearly 70 years.
For decades, most technically capable CPUs and GPUs owned by a household or small business were typically a video game console or gaming-focused PC. No other computing software required the horsepower of a game. In 2000, Japan even placed export limitations on its own beloved giant, Sony, fearing that the company’s new PlayStation 2 device could be used for terrorism on a global scale (for instance, to process missile guidance systems).2 The following year, in touting the importance of the consumer electronics industry, US Secretary of Commerce Don Evans stated “yesterday’s supercomputer is today’s PlayStation.”3 In 2010, the US Air Force Research Laboratory built the 33rd largest supercomputer in the world using 1,760 Sony PlayStation 3s. The project’s director estimated that the “Condor Cluster” was 5% to 10% the cost of equivalent systems and used 10% of the energy.4 The supercomputer was used for radar enhancement, pattern recognition, satellite imagery processing, and artificial intelligence research.5
The companies that typically focused on powering video game consoles and PCs are now some of the most powerful technology companies in human history. The best example is computing and system-on-a-chip giant Nvidia, which is far from a household name yet ranks alongside consumer-facing tech platforms Google, Apple, Facebook, Amazon, and Microsoft as one of the ten largest public companies in the world.
Nvidia’s CEO, Jensen Huang, didn’t start his company with the intention of it becoming a gaming giant. In fact, he founded it based on the belief that eventually graphics-based computing would be needed to solve queries and problems that general-purpose computing never could. But to Huang, the best way to develop the necessary capabilities and technologies was to focus on video games. “The condition is extremely rare that a market is simultaneously large and technologically demanding,” Huang told Time magazine in 2021. “It is usually the case that the markets that require really powerful computers are very small in size, whether it’s climate simulation or molecular-dynamics drug discovery. The markets are so small, it [sic] can’t afford very large investments. That’s why you don’t see a company that was founded to do climate research. Video games were one of the best strategic decisions we ever made.”6
Nvidia was founded only a year after Snow Crash—which the gaming community also quickly considered to be a seminal text. Despite this, Stephenson has said the emergence of the Metaverse via gaming is “the thing I totally missed in” the novel. “When I was thinking up the Metaverse, I was trying to figure out the market mechanism that would make all of this stuff affordable. Snow Crash was written when 3D imaging graphics hardware was outrageously expensive, only for a few research labs. I figured that if it were ever going to become as cheap as TV, then there would have to be a market for 3D graphics as big as the market for TV. So the Metaverse in Snow Crash is kind of like TV. . . . What I didn’t anticipate, what actually came along to drive down the cost of 3D graphics hardware, was games. And so the virtual reality that we all talked about and that we all imagined 20 years ago didn’t happen in the way that we predicted. It happened instead in the form of video games.”7
For similar reasons, the software solutions that are best at real-time 3D rendering come from gaming, too. The most notable examples are Epic Games’ Unreal Engine, as well as Unity Technologies’ eponymous engine, but there are dozens of video game developers and publishers with highly capable proprietary real-time rendering solutions.
Non-gaming alternatives exist, but at least for now, they’re widely regarded as inferior for real time, specifically because that constraint wasn’t necessary to them from the start. The rendering solutions designed for manufacturing or film did not need to process an image in 1/30th or 1/120th of a second. Instead, they prioritized other objectives, such as maximizing visual richness or the ability to use the same file format to both design and manufacture an object. These solutions were typically designed to run on high-end machines, rather than nearly every consumer device worldwide.
One advantage that’s often overlooked is the fact that game developers, publishers, and platforms have had to fight and work around the internet’s networking architecture for decades and thus have unique expertise as we shift to the Metaverse. Online games have required synchronous and continuous networking connections since the late 1990s, with Xbox, PlayStation, and Steam supporting real-time audio chat across most of their titles since the mid-2000s. Making this work has required predictive AI that takes over for a player during a network drop before handing back control, custom software to unnoticeably “roll back” gameplay in the event that one player suddenly receives information before another, and creating gameplay that aligned with, rather than ignored, the technical challenges likely to affect most players.
This design orientation leads to the final advantage games companies possess: the ability to create a place someone would actually want to spend time in. Daniel Ek, the co-founder and CEO of Spotify, has argued that the dominant business model of the internet era has been breaking down anything made of atoms into bits—what was once a physical alarm clock on a nightstand is now an application inside the smartphone on a nightstand, or just data stored on a smart speaker nearby.8 In a simplified sense, the Metaverse era can be thought of as involving the use of bits to produce 3D alarm clocks made of virtual atoms. Those with the most experience in virtual atoms—decades of it—are game developers. They know how to make not just a clock, but a room, a building, and a village populated by happy players. If humanity is ever to move to a “massively scaled interoperable network of real-time rendered 3D virtual worlds,” that skill is going to take us there. When discussing what he got right and wrong about the future in Snow Crash, Stephenson told Forbes that “instead of people going to bars on the Street in Snow Crash, what we have now is Warcraft guilds” which go on in-game raids.9
In the first part of this book, I’ve detailed where the term “Metaverse” and its ideas come from, the various efforts to construct it over the past several decades, as well as its importance to our future. I’ve investigated the corporate enthusiasm for this would-be successor to the mobile internet, reviewed how this confusion was and continues to be significant, introduced a workable definition that explains what the Metaverse is, and touched on why video game makers seem to be at its forefront. Now, I’ll walk you through what it will take to make the Metaverse a reality.
* Another important leader in the mobile device market is Samsung, which, unlike these other manufacturers, is 80 years old. However, it never held significant market share in the mainframe nor PC markets.
† The term “internet” is an abbreviation of “inter-networking.”
‡ It has been argued that the internet is regionalizing, most notably the Chinese internet, and to a lesser extent, that in the EU. To the extent this claim is valid, it would be due to regulatory enforcement that results in key (and required) differences in standards, services, and content.