CHAPTER 2
Second Life and the Mystery of the First Metaverse Platform

It's November 30, 2006, in San Francisco, at the Commonwealth Club of California. It's the largest public affairs forum in the United States. Martin Luther King Jr. has spoken there, as have multiple American presidents and countless luminaries in literature, science, and business.

On this night the topic is “Online Personas: Defining the Self in a Virtual World.” The panelists include the CEO of LinkedIn and Robin Harper, VP at the startup Linden Lab, developer of Second Life.

And, at the cherubic age of 22, the founder of a social network for college kids, something called Facebook.

Tech executives and Bay Area Brahmins are gathered to hear them discuss the future of online identity.

The moderator's first question, along with the panel's general framing, assumes that 3D virtual worlds like Second Life will define the future social experience online—so much so, that the panelists are even asked what kind of avatar they'd like to have in a virtual world. Mark Zuckerberg hesitates at first, then says being an avatar who looks like Cher might be fun.

No one then—not even perhaps Zuckerberg himself—could have foreseen what would transpire over the next 15 years.

It's one of Silicon Valley's greatest mysteries, and it's haunted me for over a decade:

How did Second Life, among the most well-publicized technology platforms of the early 2000s, anointed by countless experts as the coming of The Metaverse itself, fail to fulfill its grandest aspirations?

Now with the entrance of Meta, and continued interest in the Metaverse, that mystery is relevant to anyone wondering whether—as many now proclaim—immersive virtual worlds will be the future of the Internet.

If that's true, one may fairly ask, what happened to Second Life?

Because every breathless forecast about the Metaverse that you have read since 2020 was also said about Second Life between the years 2005 and 2010. Countless articles and news segments depicted Second Life as the utopian future of the Internet.

Jeff Bezos invested in it. IBM created an official corporate campus in it; Harvard and hundreds of other major schools and organizations erected virtual outposts in it. Second Life was featured in The Office and CSI: New York, with many more TV shows and movies to follow.

But it was not merely the media coverage that elevated Second Life. The very concept of the Metaverse, as embodied in Second Life, fascinated brilliant people from unexpected realms.

When he was preparing for a potential presidential run, former Virginia governor Mark Warner (now a leading senator) made a whistle stop in Second Life.

Famed inventor and futurist Ray Kurzweil confidently predicted that by 2045, we would start uploading our consciousness into virtual worlds like Second Life—and even gave a keynote address at the 2009 Second Life Community Convention.

Thomas Pynchon, the world's greatest living novelist, published a novel (Bleeding Edge) partly set in a “pixelated landscape” that eerily resembles Second Life—inviting speculation that the famously reclusive author was wandering the Linden Lab's world as an incognito avatar.

Celebrities, too, anonymously explored the virtual world. Comedian Drew Carey would occasionally tweet about his adventures in Second Life and Nylon Pinkney, his favorite avatar creator. For a tantalizing time (I'm told by a reliable source), a Hollywood project inspired by Second Life was planned to star Brad Pitt and Sacha Baron Cohen, with Pitt playing a character like Second Life founder Philip Rosedale and Cohen one of his trollish users.

Enthralled by all this attention, hundreds of thousands and then tens of millions of people signed up for the virtual world. As of this writing in early 2023—two decades after Second Life's official launch—nearly 70 million Second Life accounts have been created.

By 2007, Linden Lab had informally achieved fabled unicorn status, attaining a $1 billion-plus valuation based on secondary stock sales. Catamount Ventures reportedly sold 10 percent of its Linden Lab stock for $500 million (a source told TechCrunch in November 2007), suggesting valuation in the several billions of dollars.

Microsoft was even widely rumored to be courting Linden Lab to buy Second Life outright.

Asked about that now, Philip Rosedale says he can't recall serious acquisition talks with Microsoft but does clearly remember a senior executive with IBM saying to him, “We're just going to overwhelm you with interest. You're gonna have to basically just tell us to leave, because we'll have meetings with you all day long.” By then, IBM not only had an official corporate campus in Second Life but upwards of 10,000 employees regularly using it.

“If anyone was going to acquire us,” says Philip now, “it probably would have been IBM.”

He does confirm Linden Lab was on track back then to have an initial public offering.

“We were getting ready for it because our revenue growth and our margins were so good. We had a sales team; we had all the right qualities for a company to IPO.”

But then something strange happened, and it happened during the peak of Second Life's mainstream accolades, interest, and financial success.

The user base.

Simply.

Refused.

To grow.

Tens of millions of people tried Second Life, several hundreds of thousands every month, excited to see what the fuss was all about.

But somehow, roughly 99.9 percent of them quit, usually within the first few minutes.

At peak in 2006, the virtual world plateaued at under 1 million monthly active users, where it's roughly remained ever since.

But why? And what does the answer tell us about the future of the Metaverse?

I've talked to dozens of key insiders over the years, many of them on record for this book. (Still many more over many drinks after careening into the inevitable question, “Where did it go wrong?”)

The explanation for Second Life's current state is fascinating, frustrating, and, at its heart, fundamentally human. Far from being a simple tale of technology, Second Life's fate is best understood with the highest ideals that drove it—and the conflicts that inevitably emerged while trying to conceive a utopia out of thin air.

I'm also convinced that we will not have a Metaverse that matters until we understand how Second Life failed in that goal—and just as key, succeeded in its own way against all odds.

And the answer to what went wrong with Second Life's grandest dream starts in the middle of the desert.

Part 1: Rise and Fall

Burning Metaverse

As I wrote in The Making of Second Life, the original intention was not to create the Metaverse per se, but rather a simulated virtual world with its own ecosystem to explore. But where earlier online worlds drew their inspiration from science fiction or fantasy, a genesis of Philip Rosedale's vision was Burning Man, the orgiastic art festival held every year in Nevada's Black Rock desert.

Philip was inspired not only by Burning Man's endless freeform creativity taking shape from the nothingness of the playa, but by the social connectivity that emerged:

“I was just blown away by the fact that I was willing to talk to anyone,” Philip Rosedale told me then. “That it had this mystical quality that demolished the barriers between people. And I thought about it: ‘What magical quality makes that happen?’”

To this day Philip considers this the essence of Second Life: “I always had an intense desire to communicate, wholly, to communicate in a genuine and intimate way with others.”

The quicksilver alchemy of connecting strangers remains core to Philip's vision throughout Second Life's long evolution.

The very early versions of Second Life prior to its 2003 launch actually resembled rudimentary versions of Minecraft and Fortnite, and it's tantalizing to consider what the virtual world would look like, had it continued evolving in that direction.

“There was a time where Linden and Second Life were trending a little more toward Minecraft, where the physics and the visuals of everything will determine how everything works,” as founding CTO Cory Ondrejka remembers. (As with Minecraft, the team had also created feral creatures that would hunt down the player.)

Also guns. Lots and lots of guns:

“We had a version that was very much what early versions of Fortnite ended up being, where you could shoot at each other, but you can also put up walls and break things.”

Cory Ondrejka describes himself as the cofounder of Second Life, alongside Philip Rosedale, and he does so with some justice.

“Virtually all of the original core systems, other than the initial render, the original simulator, the original space server, all those initial versions, I wrote basically the first version as a proof of concept,” as Cory puts it to me now. Much of Ondrejka's original code still exists in Second Life today.

By “cofounder,” Cory Ondrejka also means how closely he and Philip worked together to build out their world.

“Second Life was very much a co-creation, and I don't think you would have gotten to where it was if either of us had tried to push it forward on our own,” as Cory remembers it now. “And I think the dynamic tension of Philip and I not always agreeing and bouncing off each other and challenging each other is part of what enabled us to build such an energetic company, such an amazing product, and help drive things forward so quickly during that period.”

When I met Ondrejka in 2003, he had dark hair and a way of speaking that veered between geekily animated and gravelly sardonic—a holdover, perhaps, from his early stint as a U.S. Navy officer. During the pandemic, however, his family decided a cheerful makeover to his now-graying hair was in order. So when I reconnected with him on video calls for this book, Ondrejka delivered in-depth technical and strategic analyses from under a head of rainbow-colored hair.

Developed in the early 2000s alongside Philip Rosedale and a handful of developers, these early incarnations of Second Life (dubbed Linden World at the time) came with creation tools, enabling players to instantiate building blocks—called “prims,” short for “primitives”—into the world, connecting and customizing them into virtually any conceivable object, even imbuing them with physics and interactivity, through Linden Script Language (which Ondrejka whipped up over a few days).

It was these creation tools that the startup and its investors realized were the product's key features. They also set Second Life on the explicit path of becoming the Metaverse. As I wrote in my first book:

  • [While] Rosedale and Cory Ondrejka spoke to their financial backers, a projector displayed a live video feed of Linden World, projected on the wall. Other Linden staffers were in-world, running a demonstration that the investors could watch. A few of them were using the building tools the staff used to create content. And as it went on, the investor's eyes drifted away from the meeting, and to the screen.

    On it, one Linden staffer was building a giant, evil snowman—and another staffer was busy creating a mass of little snowmen, gathered around their titan Frosty, to worship him. This, everyone realized, was what made their world unique: To build and see the results instantaneously; to share the act of creation with others; to riff off their work, and make it larger than its individual components; to be in a collectively shared collaboration with people from all over the world. That was the uniqueness that they had stumbled upon, without quite planning it; that was the key feature that would distinguish Second Life from everything else on the market.

    [Only then] did Ondrejka begin to see a connection between what they were making and Stephenson's Metaverse. “People are going to build human artifacts,” he realized. “And if you're going to have human artifacts, you're going to need to have people … which means people, which means avatars.”

Indeed, by the time I joined Linden Lab as a contract writer/embedded journalist in early 2003, Snow Crash had well ensconced itself as a reference point for Second Life as a product. On an early visit to the office, I noticed that the novel sat in a prominent place on the startup's bookshelf. When the developers discussed different aspects of the virtual world, such as point-to-point teleportation and virtual real estate, Snow Crash was often cited.

Off Game

But the rise of the Metaverse as a reference point contributed to a growing ambivalence within Linden Lab around what, exactly, Second Life was as a product.

It vaguely resembled The Sims Online, Electronic Arts’ virtual world spinoff of the classic Will Wright game franchise. Second Life was marketed along similar lines, including a 2003 appearance at E3, the game industry's signature retail conference. Second Life even launched with a leaderboard based on user-to-user ratings.

At the same time, Second Life was always presented as an open-ended world without explicit goals, quests, or missions. (“If you can imagine it, you can do it here,” Second Life's 2003-era website stated flatly. “You choose your own goals.”)

“Second Life had always tried to distance itself from the gaming universe,” as Richard Nelson remembers it. Nelson joined Linden Lab as an engineer in 2001 and would remain on staff for nearly 20 years. “It's fair to say among many Lindens, that approach precluded adding scores and gamifying the experience. In the early days, we saw ourselves as the browser of the Metaverse. Did it make sense to gamify the Metaverse?”

That perspective eventually became official policy. In 2005, David Fleck, Linden Lab's newly appointed head of marketing, announced in a company-wide email that thenceforth, Second Life would no longer be called a game. (Linden staffers back then, including me, often did so.)

“Yes, this was a messaging initiative that was implemented mostly for PR purposes,” Fleck tells me now about making the word verboten. “The press was discovering virtual worlds and would immediately categorize SL as a ‘game.’” In other words, Linden Lab faced a “‘looks like a duck, swims like a duck, and quacks like a duck, then it probably is a duck’ problem.”

Fleck also saw “an opportunity for Second Life to attract non-gamers by helping them understand that it's a place to hang out, to be social should they want that, and more importantly, not compete for some end goal, as is typical with games.”

Instead, Second Life was from then on to be described as a “virtual world” or a “platform” or both, open to varieties of use cases.

“That was to attract more creators,” Fleck tells me now. “The thinking was that more content (creators) equals attract more users equals growth.”

This helped position Second Life as a business opportunity: “Create content in SL and make big money—a message clearly targeting creators interested in tools/monetization (platform).”

It was also an attempt to frame Second Life as the “3D web” that external organizations could build on, creating their own immersive spaces, similar to how they made home pages. I mildly protested this direction in the group email thread, but as a freelance contractor, I had little lobbying heft.

“I'm still trying to remember exactly when did I first have that thought, you know: ‘Second Life is not a game!’” Philip Rosedale told me in 2022.

Now that he’s in his 50s, Philip's once-blondish hair leans toward platinum but is still cut with a rakish bounce, his boyish features faded somewhat but still punctuated by piercing light blue eyes and photogenic looks, which helped buoy his rise as a Silicon Valley celebrity. (The New Yorker more succinctly described his current appearance as “Danish movie star.”)

Looking back, however, he says the distinction wasn't about game mechanics as such, but how he wanted people in Second Life to engage with each other: first and foremost as other people. (Or as Kant put it, describing the best moral path, treating others as ends in themselves, not as means to an end.)

Philip again: “The point that I was trying to make, mostly, when I said, ‘Second Life is not a game,’ was merely to say that the nature of Second Life, no matter what it turned out to be in the long term, or no matter what it was that it would take to get a billion people in there, was going to be a lean-forward experience, where a primary part of the experience is engaging with other people who are nearby.”

Second Life could not be a game, in other words, if it were to capture that transcendent sense he had of being able to meet people from all walks of life in the Nevada wasteland.

Metaversed

Floundering with little user growth after its official launch in June 2003, Second Life finally hit escape velocity when it explicitly adopted an essential facet of the Metaverse: integration with the real-world economy.

Around the start of 2004, the company announced that the virtual world's currency, Linden Dollars, could be bought and sold on the open market for real cash. And Ondrejka drove the effort to implement what was a revolutionary concept at the time: enabling users to own the underlying intellectual property rights to anything they created with Second Life's building tools.

“That was me on the phone with lawyers getting laughed at and the lawyers saying, ‘You can't do that.’” But Linden Lab did eventually do that, formulating an IP rights policy that YouTube and other user-generated content platforms would come to adopt.

From this point onward, Second Life users could, and quickly did, create small businesses from their virtual world content. This in turn grew the world of Second Life in a way that greatly amplified Linden Lab's own work on the platform.

“[A]t that point we had like, 100,000 people [playing] and we computed that,” Cory Ondrejka remembers. “Ten percent of the people were actually doing awesome things in the world. That'd be like having a 10,000-person company. Because when you're a 150-person company, a 10,000-person company is unimaginably large.”

Major press coverage of Second Life, often by tech reporters who were fans of Snow Crash themselves, began to gather momentum into 2006, culminating in a feature story on the cover of BusinessWeek—which catapulted into even further coverage. Strikingly, most of this overwhelmingly positive media attention was organic, generated by reporters and media outlets themselves.

“It was kind of terrifying,” Catherine Smith, head of Linden Lab communications back then, recalls now. “I remember [a Linden Lab engineer] saying to me one time, ‘Can you just please make the PR stop? And I'm like, ‘Sorry, that train has left the station. I'm not pitching this anymore. They're all coming to us.’”

Charity Majors, an engineer who helped keep the glorious hairball of Second Life's complex backend operations running during its peak years, saw this rise of media attention play out in sharp spikes of new users.

“The main one that I remember was, of course, the episode of The Office.”

An entire subplot of “Local Ad” (Season 4, Episode 9) was devoted to Dwight and Jim's nerdy explorations of Second Life. The show producers even interviewed Linden Lab staff while preparing the script, and so managed to perfectly capture in its dialog the fundamental dilemma of the virtual world's identity:

  • Jim: You playing that game again?

    Dwight [with serene confidence]: Second Life is not a game. It is a multi-user virtual environment. It doesn't have points or scores. It doesn't have winners or losers.

    Jim: Oh, it has losers.

Charity and other team members were in Linden Lab's own office as the show aired, awaiting the deluge.

“And there was just a flood, a hoard of people logging on and everything was blinking red,” she says. “And we're just running around batting shit left and right. It was nuts.”

Interest in Second Life was now so fervent and sustained, and so global in scope, Charity remembers Linden's engineering team collectively realizing over a team meal that the virtual world had effectively become too big to fail.

“One of the fun questions was, ‘If Second Life ever went completely down, do you think it would ever come back up? Because, the load profile of turning things back on is what's high.” Were Second Life to go offline, in other words, it would be hit by an endless cascade of frantic users around the world all repeatedly trying to log back on, overtaxing the system in the process.

By then the virtual world was thriving, with a population roughly the size of San Francisco’s and a total landmass approaching that of Los Angeles.

In naked economic terms, virtual land sales and maintenance fees translated into real and rapidly growing revenue and represented thousands of Second Life users paying Linden Lab millions of dollars every month to own a plot of digital real estate—from a small homestead costing a few dollars to a private island costing hundreds to a mini-continent priced at thousands every month.

In a Chinatown restaurant around that time, Cory Ondrejka regaled me and other Linden staffers with a plan to turn Linden Dollars into an electronic currency, for use to buy any kind of goods and services, even the sort that you might hold in your hand.

“We recognized that here [with virtual currency] is a way that you lower friction, lower cost transactions, and do them in a very consistent and native way within the world,” Ondrejka explains now. By then, it was possible for users to buy and sell Linden Dollars among each other, and some of them were already experimenting with them as a means to buy physical items like computer peripherals.

Years before Bitcoin existed, in other words, active plans were already underway to make L$ the first broadly used digital currency. Bolstering that economy, real-life companies starting with Coca Cola began to establish an official presence in Second Life, which further excited the media.

Beneath the surface of that buzz, something more rare and precious quickly took shape.

Emerging Culture, Rising Creators

The breadth and ambition of creativity in Second Life was almost immediately astounding. Looking back at it now, I see it as a microcosm of what the Metaverse in full flower could eventually be.

Inspired by Robert Heinlein's classic science fiction short story “—And He Built a Crooked House,” a mathematician created a mind-melting four-dimensional tesseract house; working in her spare time, a game developer built Svarga, an island with a self-sustaining, fully simulated ecosystem; and a full-fledged roleplay community created the Wasteland, a sprawling tribute to various postapocalyptic pop culture classics.

This virtual expression wasn't confined to pixels. Logging in from Beijing, renowned conceptual artist Cao Fei erected an entire city in Second Life, sold virtual real estate deeds to it at Art Basel, and eventually brought the project to New York's MOMA. A full 14 years before rapper Travis Scott's avatar would grace the stage at Fortnite, platinum-selling artist Suzanne Vega performed live in Second Life—one of many performers to start or punctuate their careers from a virtual world venue. Helen, a multimedia producer in Colorado, would regularly transform her avatar into a dignified old woman dressed in old European finery, then hand the computer controls to her aging mother, Fanny Starr, a Holocaust survivor, who'd then tell groups of gathering Second Life users from around the world about her time in Auschwitz.

And for many creators working in Second Life, what they made also transformed who they were offline, often for the better.

Jeff Berg (whom we met in the Introduction) was among the first to gain fame as a metaverse artist. Through his avatar persona, AM Radio, Berg tells me, he was trying to evoke a sense of the sublime in the Metaverse, creating “spaces that are stable, unmoving, intangible, untouched and almost untouchable, unchanging, deep-rooted; places that might be points of reference, of departure, of origin.” That's even evident in the names he gave these Second Life spaces: The Far Away (see Figure 2.1), Towards the Sky, The Quiet.

“All of my work, art or not, has a hope draped across it,” as Jeff Berg puts it now. “That we remember always, despite all the wonders of the advancing technology around us, that our collective definition of who we are and the world we design for ourselves is rooted in the wonder of the sublime our hearts prefer.”

As Jeff Berg became famous in Second Life as AM Radio, and as his work impacted people's lives, their response to him began to change Berg himself.

Before Second Life, he had low self-esteem at work and resisted advancement opportunities for that very reason. But now that people in Second Life were clamoring to meet AM Radio in-world, and because he felt obliged to meet them, he also found it easier to engage with people he met in the real world and assert himself more. At IBM, he tells me, he became more outgoing, more willing to take on leadership roles. Before becoming a famous avatar, that would have been impossible.

Photograph of AM Radio/Jeff Berg in The Far Away in Second Life.

FIGURE 2.1 AM Radio/Jeff Berg in The Far Away in Second Life.

“Being AM Radio absolutely got me past a barrier,” he says, “because I had to.” Self-esteem and confidence thrived, and even years later, remains: “I will talk with anyone,” he says. “I cannot be a fly on the wall.”

Jeff's transformation echoes what I've seen in countless Second Life denizens; so much so, I gave it a name: Mirrored flourishing, the concept that positive contributions to the virtual world can and should have a positive impact on people in their offline lives—and vice versa.

For Gizem Mishi (whom we also met in the Introduction), the virtual world helped restructure the course of her entire life. She first entered Second Life from her flat in Turkey to attend a Skye Galaxy concert. This is where she first conceived of Blueberry, her virtual fashion brand. While Skye was singing, she also met an American man with a nondescript avatar, who mainly logged into Second Life to play in a user-run virtual football league.

“We ended up talking for eight hours straight with my broken English,” Gizem tells me. “I still think it was my polka dress that lured him in.”

Offline, Gizem eerily resembles the avatars she customizes to model her virtual fashion—20s, long brunette hair, full lips. She still speaks with a lilting Turkish accent. (Her name, by the way, means “mystery.”)

She talked with this avatar jock in Second Life for three months more. And then Brandon McDuff, a former college football player from North Carolina, flew to Turkey to meet Gizem herself.

A year and a half later, he moved to be with her in Turkey. And then Gizem became Mrs. McDuff.

Blueberry was not her first brand, it should be said, for even though she founded it while still in college, by then she had already worked for two successful online startups, beginning at the age of 15, and despite logging in remotely from Izmir, a Turkish city off the Aegean sea, worked her way up to a founder's role at Peanut Labs, a San Francisco-based game monetization startup.

But then in 2016, a faction within the Turkish Armed Forces launched a violent coup d’état against President Erdoğan, which was somewhat inconvenient for the deadline for Blueberry's latest release in Second Life.

Amid the rumble of explosions outside and choppers striking government buildings in Istanbul, the Internet and electric power in her home struggled to stay online; Gizem's husband and her father set up a power generator to maintain her connection to the virtual world. Outside her window, bright orange tracer rounds lit up the sky.

“Places were getting mortared,” she recalls now, voice drifting off. “It was really, really … it was something.” Gizem managed to make her Arcade fashion event deadline at 4 a.m., uploading all her items just before her connection to Second Life was lost.

In Snow Crash, the hero's story culminates in a high-speed chase within the virtual world, a struggle over a data file transfer, while at the same time paramilitary choppers and various combatants battle it out in the real world. Metaverse history doesn't repeat itself, but it does often rhyme.

But the coup was also bad for virtual business. In the aftermath, Erdoğan locked down Turkish access to Internet services that managed payments online, including PayPal and, for a time, Second Life itself.

The virtual fortune Gizem had worked so hard to amass now drifted in the purgatory between the virtual and the real.

Linden Lab's CEO at the time personally interceded, helping create a way for her to repatriate these funds. And after all this high-stakes international intrigue, her husband insisted they move back to the United States. Gizem's mentor, Prosper Nwankpa, helped set up a business and bank account based in America. The Metaverse is only as functional as the human infrastructure beneath it.

Which is why Blueberry is not only incorporated as a business in the United States but also has a brick and mortar office in Raleigh, North Carolina, a short drive from their shared home.

At Second Life's peak, stories like these seemed infinite and ever varied, evidence of a thriving, fully realized Metaverse culture that seemed even more vital than anything Snow Crash once depicted. For companies and organizations interested in setting up a beachhead in Second Life, it ratified their sense that this was, in fact, The Future.

But churning beneath this shiny surface, little noticed at first, a growing danger threatened it all.

Failure to Retain

“Second Life had been growing very quickly, but that growth had been masking retention problems,” Ondrejka remembers now.

“[It] was like 2008 where the sigmoid curve, the growth started to roll off and that was unambiguous,” as Philip Rosedale puts it from the vantage of 2022. “We also knew something that had been true all along, but we thought we would make more headway on, which was the percentage of people who signed up who actually stayed around for, say, a couple months, was really low. Like in the single digit percentages. And I think we always thought that we could fix that.”

“People just bounced off it,” as longtime Linden Lab engineer Richard Nelson remembers it. “The complexity of the user interface, the learning curve.”

Second Life's user experience was only one among many obstacles, and in truth, few prospective users even managed to reach the first-time installation process.

Another Linden veteran once told me (as my jaw steadily dropped) that most of these millions of new signups quit before the actual Second Life program had even completed downloading on their computer.

It's humbling how few consumers are accustomed to installing new programs onto their PCs—and how impatient they become if they're not instantaneously entertained. As it happens, the only key consumer group which is accustomed to regularly downloading large new programs, especially those with 3D graphics, are PC gamers.

But Second Life was not being marketed as a game.

The few would-be users who were able to install the client and log into Second Life itself faced an even more daunting challenge: What do I do now?

Since Second Life wasn't a traditional game, with no explicit tasks, goals, or scores, the vast majority of them were too intimidated, overwhelmed, or frankly just too bored to go any further.

“When you enter Second Life, you either create content or consume content and experiences,” as David Fleck puts it. “The majority of users are not content creators—they either lack the skills or have no interest in building content.”

Fleck drove Linden Lab's marketing of Second Life at the time and faced the mind-boggling challenge of positioning a product with no explicit purpose.

“If you aren't creating content, you need to be interested in consuming what others have built, or you need to be ready to engage with other community members to socialize with them.” But many if not most people are comfortable walking into a public space and socializing.

“Second Life is an open canvas that allows anyone to decide what the purpose will be to address their own needs. But that's like asking people in the real world, ‘If you could do anything, what would you do?’ For most people, that's a tough question to answer because they don't know the answer.”

Second Life, in other words, confronted nearly everyone who encountered it with an uncomfortable truth about their lives that they had not yet faced. And then charged them money to face it.

Second Life originally launched with a subscription fee, but Fleck convinced Rosedale to shift Second Life to a freemium model, which skyrocketed new user signups but did little to improve retention.

This savage attrition was difficult to perceive on the outside. Business reporters unfamiliar with Internet metrics happily reported that Second Life had “over two million users,” when that number actually referred to registered users (that is, people who had created an account), not monthly active users (that is, people who actually returned after registration).

Respected Internet academic Clay Shirky, among the few early skeptics, suggested this hypothesis in a 2006 essay:

  • I suspect Second Life is largely a “Try Me” virus, where reports of a strange and wonderful new thing draw the masses to log in and try it, but whose ability to retain anything but a fraction of those users is limited. The pattern of a Try Me virus is a rapid spread of first-time users, most of whom drop out quickly, with most of the dropouts becoming immune to later use.

I wrote an anti-Shirky diatribe in response back then, convinced he was underestimating Second Life's growth potential. After all, if Bezos was an investor and major companies were rapidly setting up an “official HQ” in Second Life, surely this retention problem would work itself out.

I later found out that Linden Lab itself was also blinded by the problem back then. Cory Ondrejka names it as one of the core issues that could have changed Second Life's fate:

“I think Clay wasn't 100 percent right, but we weren't 100 percent right, either. And I think we could have gone harder and treated as existential the fact that people were bouncing off Second Life as hard as they were.

“And instead, I think we were [thinking] ‘Well, they'll reengage, and maybe the data is a little fuzzy.’”

The company desperately flailed at staunching the loss.

“[We tried a] million different changes to the onboarding experience,” Philip Rosedale remembers. “We tried, what, probably 10 or 20 different versions of the new user experience and twiddled around with different things, like how soon you met somebody else, what sequence of instructions you were taken through.”

The tiny percentage of people who were familiar enough with 3D online worlds to download the client and successfully log in faced another problem: While Second Life looked like an MMO, it didn't perform like one.

Because unlike standard MMOs, the entire 3D experience of Second Life is streamed onto a user's computer. This made it possible for users to create and update their environments in real time. That also meant that reality itself would often seem to sputter, pause, or appear as an undifferentiated blob. Performance was generally egregious for people without a desktop PC and dedicated broadband line—and they were a steadily shrinking market in the era of laptops and Wi-Fi, let alone smartphones.

In the end, hardly anyone got through that gauntlet. Those tiny few who did stepped into something wonderful.

“Once you got past that hump, the possibilities were just endless, the things you could build or do. It was sticky as fuck,” as Charity Majors observes. “The people who got into it, and learned to use it, and made connections, would never leave. But getting them over that hump is something where I feel like there was a little bit too much ideological purity standing in the way of our success.” (More on that in the “Choose Your Own Inertia” section of this chapter.)

This growing catastrophe quietly metastasized even as Second Life was still being feted in the media, while a small but dedicated user base of some 600,000–800,000 (when you subtracted the monthly sign-up churn), the chosen few, were happily creating a thriving virtual community.

But as Ondrejka says, it was an existential threat to Second Life's grandest plans.

I first heard an inkling of this trouble in 2007 from an insider I deeply trusted.

“They're scared,” she told me, referring to Linden Lab. “They don't know what to do.”

Still enjoying the euphoria of writing The Making of Second Life while being invited to speak about the virtual world in far-flung cities around the world, I was shocked. (But not shocked enough, in retrospect, for her words to fully register.)

Open to Closing

Linden Lab made multiple major attempts to grow the user base. Among the first was a substantial revamp of the user interface. Long in development, informally dubbed “Viewer 2,” it resembled a web browser for 3D content. (Again, not a game.)

“It bombed in a big way,” says Richard Nelson.

In a cruel irony, it was an idealistic move by the company that largely condemned Viewer 2.

In 2007, Linden Lab open sourced the code to the Second Life viewer, hoping this would encourage the creative community to improve on and customize it, which would in turn foster more creativity.

“While it is clearly a bold step for us to proactively decide to open source our code,” as Cory Ondrejka said in the official announcement, “it is entirely in keeping with the community-creation approach of Second Life.”

This move did inspire the development of many user-created third-party viewers (known as TPVs), versions of the Second Life software with additional or enhanced user customization controls.

It also placed the company in direct competition with its own community.

Vexed by the user interface changes implemented in Viewer 2, the established user base mutinied, rejecting the company-made viewer and preferring a TPV that was optimized for veteran players.

This made it even more difficult to encourage new users to stay. They'd create an account, download the official software, log into Second Life, and encounter established users who were using a TPV with a completely different user interface—in effect, a whole other perceptual reality.

“People come in and the first thing [veteran players] tell you to do is download another client,” notes Nelson.

By 2012, a Linden Lab engineer reported that the official Second Life client was the third most popular viewer among the user base. This effectively made new user growth nearly impossible.

Linden Lab open sourced its software to encourage creativity and innovation among its users. In retrospect, it's no surprise that these users leveraged this openness to serve their own immediate desires, even at the expense of the world's long-term success.

It illustrates the double-edged power of a metaverse platform: At some point, the user community's contributions to the platform become so popular and seem so indispensable, they become essential to the platform itself. Sometimes leading to painfully unintended consequences.

While attempts to grow user retention floundered, a new opportunity emerged in 2007 with the launch of the iPhone. Selling 20 million units in its first two years, the iPhone's instant popularity augured a new paradigm for online activity. By 2008, rumors began swirling that Second Life would join that transition with an official iPhone app. Surely the company realized it should?

If only that move would get enough love.

Love Machine and the Tao of Linden

Inspired to create Second Life in part by the democratizing openness of Burning Man, Philip Rosedale brought similar idealism to Linden Lab, the company building it.

“Philip, I think, had this really amazing idea, which was: I want a company without managers,” as Cory Ondrejka explains, characterizing the vision thus: “‘I want just a company that can scale indefinitely. And everybody just kind of does what they want, but what they want is the right thing.’

“It's why he loved ideas like the Love Machine.”

While the name may evoke a glowing pink sex toy with mechanized appendages, the Love Machine actually was a computer applet inspired by The Future of Work author and MIT professor Tom Malone. It came out of Philip and Cory wanting a way for employees to send each other praise for accomplishing specific tasks and projects, along with a nominal gift of one U.S. dollar. Cory called it tipping; Philip redubbed it the Love Machine.

In effect, Linden Lab had taken something like the user-to-user rating and leaderboard system they gave to early Second Life players, and applied a version of that to themselves—to help run the company. In theory, the Linden Lab employees who earned the most Love would, without any bias from management, get warmly recognized by their own peers as the most important contributors. The aggregate data of all this Love (to whom and for what) would also give company leaders insights into the kind of projects considered most crucial.

The Love Machine did have its virtues.

“I fuckin’ love it,” as Charity Majors puts it. “Just that pattern of making it fast and almost frictionless and easy to just celebrate people and shine a light on the importance of parts of the company that don't get a lot of love—so good.” After leaving Linden Lab, Charity would even go on to implement a version of the Love Machine in Honeycomb, a successful observability startup she cofounded.

Like the employees of many tech companies before the dawn of Slack, Linden staff often communicated with each other via Internet Relay Chat. To make it easier to use, one staffer automated the Love Machine, turning each Love nomination into a clickable link in this group chat. Pile-on Love giving soon became the norm.

“We'd created an incredibly powerful viral loop for ourselves,” Ondrejka wrote in a 2022 postmortem. “The bummer, of course, is that it made The Love Machine way less effective.”

Not all employees were highly active on IRC, so they largely missed out, while employees who were on IRC could see which employees were receiving the most Love: “[That] emphasized—and made public—the haves and have-nots among recipients, which hadn't been the intention.”

And now with hundreds of people on Linden Lab staff participating, those nominal one dollar tips became pretty substantial rewards. Catherine Smith remembers getting hundreds of dollars in Love tips when Second Life was positively featured in the media. She also remembers a Linden Lab staffer having to walk to the company's bank to collect the latest tranche of Love awards—and trudge back to the office through downtown San Francisco carrying $10,000 in cash. (At that point, Love was implemented into staffer's paychecks.)

In Cory Ondrejka's eyes, the Love Machine was not working as a management guide.

“Kind of like [Second Life] leaderboard stuff, it worked in a way that was self-reinforcing in ways that weren't necessarily very useful.” Were people judiciously giving Love to genuinely important tasks and personnel, or caught up in the moment, promiscuously throwing their Love around?

Despite its arguable usefulness, a number of other Silicon Valley companies have adopted a variation of the Love Machine. “It's kind of amusing to see so many of them have the same quirks that Linden had,” Cory notes impishly. “Google still uses a version of it today.”

Cory Ondrejka now believes the Love Machine led to the final company rupture that helped seal Second Life's fate.

“The Love Machine was one of my attempts to support Philip's goal of minimal management, growth without hierarchy. It didn't work, but lots of people thought it did (or at least better than it really did), so it made [my] debates with Philip even harder.”

It was easier to discount a particular proposal raised by Cory, if there wasn't enough Love to back it up. “The Love Machine rewarded people for doing things that helped you. It didn't help focus people on strategic needs.”

Whatever the Love Machine's problems, they were compounded by another idealistic management principle: Linden staffers voted on company tasks and projects through JIRA, a software development tracking service popular in Silicon Valley but not necessarily intended for running an entire company.

This approach was expressed in the company's official principles of corporate culture (cheekily dubbed “the Tao of Linden”), as: “Choose Your Own Work.”

That Tao edict had considerably more problems than the Love Machine.

“It's crazy that, you know, I was expected to vote on an engineer's project and they were gonna vote on mine,” as former PR head Catherine Smith puts it. “It's kind of like the whole ‘work on what you want’ part of the job felt like it had no application to me whatsoever. Because I gotta deal with shit whether I want to or not.”

As the company rapidly grew in size while the entire Internet ecosystem changed, that level of intransigent shit quickly became un-dealable.

The Unmaking of Second Life

Linden Lab's democratized work culture took on a different cast as the company grew to many times its original size.

“[W]hat always happens with cultures, as companies grow, is things get encoded as bumper stickers, but the context gets lost,” as Cory Ondrejka puts it now. Staff were given much freedom on which important tasks they would work on, “But it was never just ‘choose your own work,’ it was: ‘Choose your own work from the most important things to work on.’

“But as you then grow, you lose that connection back to the original idea. And so because of that, Linden at 350 people was operating in a very disorganized way… [It's] really rough for being able to make certain rapid moves, particularly with a lot of pressure on parts of the company, around scaling.”

All this happened in the growing shadow of social networks like Facebook (launched in 2004) and Twitter (launched in 2006) and rabid enthusiasm for the iPhone (launched in 2007). In this crucial moment, Second Life could have pivoted, shifting its focus to be more integrated with social networks and increasingly toward smartphones.

But it is difficult to convey a company's existential crises through the language of Love and where work projects were voted on.

“Or put differently,” Cory Ondrejka tells me, “massive pivots—massive change—needs coherent vision, leadership, and direction from the top, plus the capacity to act on change throughout the org. A Linden without Philip and me wasn't getting the leadership, so nothing the Love Machine could do would help.”

Because at that point, Cory's arguments with Philip around how to grow the company grew more heated.

“I'm bashing against Philip with, ‘We need to be a little bit more organized and a little bit more actually structured, even if it's just structured in terms of priorities … so that folks aren't trying to solve from an infinity of design space every day.’”

Linden Lab had made a virtual world that was so vast with possibilities, most people who visited it were too overwhelmed to choose what to do. And then driven by similar idealism, they had created an open corporate structure for themselves that left most of the staff similarly adrift.

Including, ultimately, the founders themselves.

Philip and Cory's clashes culminated around Thanksgiving of 2007.

“After months,” says Ondrejka, “of Philip and I meeting for many hours a week, effectively in marriage counseling, trying to find a path through our disagreements.”

Philip Rosedale finally sent Cory Ondrejka an email announcing his decision to fire him.

“It hurt at the time; it was terrible,” Ondrejka says now. “And it was aggravating. And made me angry.” He tried to exit gracefully, but at Linden Lab, “people were really unsettled.”

Cory's departure from the company impacted even me, though I'd been away from Linden Lab for over a year when it happened. Invited to the company's steampunk-themed holiday party very shortly after Ondrejka's defenestration, I vividly remember seeing Philip in a tuxedo with Victorian trim, alone at the bar, sadly staring into the middle distance.

The effort of that epochal restructuring seemed to take a toll on Rosedale too; in subsequent months after Ondrejka's departure, several Linden staffers tell me he seemed detached, melancholy, “moping.”

“I remember he came back from Burning Man, and I see Philip kind of put his head down in his hands,” as Catherine Smith recalls it now. “And I thought, ‘He's just not happy here. Now it [involves] running a big company with 300 people working there, and they were solving really hard problems. It wasn't really fun stuff anymore.”

So it was not entirely a surprise when, in March 2008, Philip Rosedale announced that he himself was stepping down as company CEO, transitioning to chairman, moving away from direct company-wide management to focus on overall development and strategy.

“We didn't know how to make [Second Life] better,” as Philip Rosedale remembered it in 2022. “We didn't know how to fix some of the problems we were seeing. That made me feel like maybe I wasn't the right CEO for that role, you know?” (One suspects Linden Lab's board and investors had similar pointed concerns.)

“Like our willingness to question everything was sometimes a liability because it could almost unglue everything … [so] I left. I felt like maybe we should try getting another CEO because that was a good experiment.”

But this move left Philip and Cory's team at Linden Lab, as talented as it was, rudderless on a day-to-day basis while facing crucial, platform-level decisions.

“Philip and I had a dynamic where we pushed each other, how we disagreed about things, where we generate ideas,” Cory reflects. “We were good at driving forward. So we both leave and you have different leaders who certainly brought interesting ideas and tried really hard to figure out what to do next. But it was a bad moment to stumble.”

Second Life's cofounders actually departed around the peak of Second Life's financial success. In the wake of Philip Rosedale's exit in mid-2008, I calculated that the company had a run rate of $100 million, of which it was clearing roughly $40–50 million a year in profit. (A well-placed insider later confirmed to me that my estimate was roughly accurate.) Meanwhile, consumer and media interest in the virtual world was still quite high.

Flush with money and popular awareness, all Second Life needed was new leadership who could grow the user base.

Linden Lab's board sought and hired one of the most qualified people to be the new CEO. And then another. And maybe yet another.

Cavalcade of CEOs

Mark Kingdon, a digital advertising veteran, took the reins from Philip Rosedale with a clear vision for remaking Second Life's first-time user experience:

“We need to give them a really simple and intuitive interface, and as their capabilities grow and their desire to create content and own land and transact grows, we need to elegantly release those features to them, rather than hit them with 350 menu choices their first moment in-world,” he told me shortly after ascending the Linden throne.

Kingdon also carried forward an ambitious partnership with IBM to make Second Life interoperable with OpenSim, an open source version of the virtual world, and also to make Second Life a viable product for hosting business meetings.

Yes, 12 years before a global pandemic would make the idea of avatars hosting corporate gatherings in a virtual world seem somewhat appealing, Linden Lab attempted selling that vision to major companies, colleges, and other large organizations.

But virtual meetings attracted few steady customers, while the interoperability project was abandoned (more on that in Chapter 8). There were improvements to Second Life's user interface, but as noted, they were actively rejected by the established user community.

After a stint of two years, Mark Kingdon stepped down as CEO in 2010.

At the insistence of Linden Lab's board, 30 percent of Linden Lab's staff were also laid off. The Second Life community created a virtual memorial to these laid-off Lindens, each of their avatar names embossed on a black marble tombstone.

It's hard to convey my own personal sense of simmering unease during this time. By then I couldn't picture a professional life apart from Second Life, surely not in an era with simple-minded Facebook games popular at the time, or rudimentary virtual worlds for kids, such as Minecraft and Roblox. (But more on that in Chapter 4.)

Second Life's floundering also felt like a betrayal of the core user base in all their quirky but soaring creativity. Would no one else stroll through the golden fields of AM Radio or listen to the velvety blues licks of Mr. Charles Bristol? And who else would emerge from a cradle of human civilization to become a self-made millionaire, following in the steps of Gizem Mishi?

I had hoped that Kingdon's successor, Rod Humble, a highly respected game designer with executive roles at Sony Online Entertainment and Electronic Arts, including work on The Sims franchise, would steady the Second Life ship. By then I'd become convinced that the company's unwillingness to position Second Life as a game first had condemned it as a consumer product.

But while the Humble administration did implement some game-like features in the user interface, these only alienated the established user community, driving them further into the arms of user-created third-party viewers.

By 2012, a Linden engineer reported that the official Second Life viewer was used only by “a minority” of its own customers. It no longer mattered how user-friendly the official software had become (and it had not), because it effectively segregated new users from the locals.

Tom Grant, analyst with the esteemed firm Forrester Research, even gave this phenomenon a trenchant name: The Iron Law of Oligarchy. Writing in 2010:

Second Life is an extreme case of how you can develop a very happy group of customers, and still fail miserably at reaching a wider audience. Some businesses are comfortable with that outcome, as long as the customer base stays loyal, and the business stays profitable. Most would be terrified to discover that their best customers are, in subtle ways, holding them back. I can't say for sure that the Second Life notables are the reason why the UI is still klunky, and the useful content is hard to find, but I definitely have my suspicions.

The Second Life community could have been cajoled and inspired to adapt a new user interface, but Philip Rosedale was no longer there to galvanize them. (And again, he was adamantly against a game-like makeover.) For that matter, few Linden staff from its earlier days were still with the company, and that further eroded the company's relationship with the community.

“The dedicated engineers we once had on staff when we cultivated a conducive culture were no longer in existence in the later years,” Kona Gurion tells me. A longtime QA manager with Linden Lab, Kona was originally recruited to the company when he was a Second Life user (a fairly common practice in Linden's early years, when dozens of employees were brought over from the user base).

“We no longer hired as diligently and with the level of scrutiny that we did before 2011, when candidates interviewed through a varied panel of people of no less than five staffers.”

The goal of that, as Kona put it to me bluntly: “Hiring people who actually give a shit about the product.” But somewhere around this time, Kona suggests, creating Second Life became for many Linden staff just another job.

Rod Humble's tenure spanned from 2011 to 2013. By then, tens of millions of dollars had been spent to grow the user base.

But Second Life's population steadfastly remained some 600,000, give or take, a small enchanted city with an eccentric but charming citizenry with countless wonders behind its walls—deep in a burning desert that few dared cross.

Third Life

Ebbe Altberg, a veteran of Microsoft and Yahoo!, was appointed Linden Lab's CEO in 2014. A stocky, voluble Swede, Ebbe seemed from the start fully confident he could bring the virtual world into the mainstream.

“I've been fascinated with [Second Life] from the beginning,” Altberg told me early in his tenure. “I don't mind those challenges, and I think we can fix this. I think there's things we can do that can make it broadly appealing, and then come back to the world….”

Ebbe soon unveiled a key piece in that plan to find a broad audience, first put into development during the Humble era: an entirely new virtual world.

First teased in June 2014, Sansar became Linden Lab's last hope to create a modern metaverse platform that avoided all the mistakes of Second Life's past.

“We believe that there is a massive opportunity ahead to carry on the spirit of Second Life while leveraging the significant technological advancements that have occurred since its creation,” as the official announcement read.

Notably, Richard Nelson tells me, Sansar was conspicuously developed without any use of the Love Machine.

Ebbe's confidence was infectious, rekindling Linden Lab's waning morale.

“We are still the shit when it comes to virtual worlds,” he told his employees during an all-hands meeting. That assertion was promptly emblazoned onto T-shirts that staff would sometimes wear.

But Sansar stumbled sharply from nearly its very first moments.

Only a few months earlier, Facebook had purchased Oculus Rift, a VR startup that first attracted wild interest as a Kickstarter, for $2 billion. The sale convinced Linden Lab that VR was the next coming trend.

“There was a lot of excitement around it,” as a Linden engineer told me, “and it seemed to have a natural fit to a metaverse. And if living in it is appealing, VR would be an important part of it.”

But it's difficult to develop a metaverse platform that is both optimized for smartphones (Linden Lab's announcements promised a mobile version of Sansar) and for VR headsets, which have high demands on graphics performance.

Faced with either option, Linden Lab positioned Sansar as a VR product at launch.

In retrospect, Facebook's purchase of Oculus put Linden Lab on an ill-fated path it could not escape. Which is somewhat ironic, because a Linden Lab veteran at Facebook helped make that purchase possible. (But more on that in Chapter 3.)

But Ebbe Altberg cannot be fully blamed for focusing so fully on VR. After the Oculus acquisition, it was widely accepted wisdom in Silicon Valley that virtual reality was destined for mass market success. In 2014, Philip Rosedale himself was so bullish, he predicted there would be 1 billion VR headset owners by 2021.

“It was all hot and hip and interesting to talk about VR, so we kind of rode that a little bit,” Ebbe acknowledged when I sat down with him in 2019. By then it was abundantly clear that VR headsets were selling nowhere near the stratospheric numbers that Valley elite once promised. (As of early 2023, the actual number of VR headsets on the market that can run a metaverse platform is decidedly not a billion, but roughly 20 million.)

While there was some early internal talk about enabling Second Life users to bring at least some of their content to Sansar, the technical challenges made this impossible. And in any case, doing that threatened to bifurcate the Second Life user base.

“We were always afraid of splintering the community. We didn't want to poach users [with Sansar],” Richard Nelson tells me. Instead, Linden Lab hoped to grow a new user base from Sansar, with the established SL community occasionally visiting its younger sibling.

Sansar struggled to find a new community apart from Second Life. And by the time it launched to open beta in 2017, several alternatives had far outpaced what either platform offered in terms of ease of usage and popularity.

IMVU, a virtual world with notably janky graphics and dismissed (if not ignored) by Linden Lab as an also-ran competitor when it launched a year after Second Life, now had millions of monthly active users. (Thanks, for the most part, to creating a version of the software for smartphones in 2014.)

Roblox, launched in 2006 at the peak of Second Life's success and therefore short-shrifted as a kiddie virtual world, had well over 50 million users in 2017.

Minecraft, launched in 2010 with a striking resemblance to Second Life's earliest incarnation, had by 2017 sold well over 120 million copies.

Sansar was sold to outside financial interests in 2020.

That same year, Linden Lab itself was bought.

By 2021, “Second Life” was displayed on the website of Waterfield Group, an investment firm that buys companies that earn a decent revenue and helps keep them profitable into perpetuity.

When I first visited the Waterfield site, I stared for some time with a dull resignation at where the first metaverse platform to reach the popular imagination was now displayed: as a portfolio holding listed beneath a company that serviced software for pumping oil and gas.

But the Waterfield move helped secure Second Life's longevity. Linden Lab's investors were finally satisfied with a liquidity event. (By then, some of them had been expecting this for over 20 years.) In Waterfield’s care, the virtual world is likely to keep running as long as its established community keeps it profitable.

Only later did we learn that Ebbe Altberg helped make this move possible even while racked with unimaginable pain.

Richard Nelson was at the all-hands meeting when Altberg announced to his team that he had cancer. It ended in a group hug of Linden staff encircling Ebbe.

“We all really liked him, and you could see it taking a toll,” Nelson remembers. “It was painful to see him fade out.”

When I first met Ebbe in person, in early 2019, he was gaunt and pale but just as high-spirited and ebullient as always. The cancer didn't cause his energy or enthusiasm to flag, even as bouts of chemotherapy ground him down.

One insider told me that Ebbe's illness gave him a serene confidence in the arc of his life course. He had a world to care for, and a community to secure, even to the very end.

When a military coup in Turkey strangled Linden Dollars at the border, for instance, threatening to disrupt a Second Life creator's livelihood, it was Ebbe who personally reached out.

“I will cry talking about this,” Gizem Mishi remembers now. “Ebbe himself responded, we had this two-hour-long talk and—may he rest in peace—they created a safe way for me to still work.”

Ebbe ran Linden Lab when the COVID-19 pandemic hit in 2020, sending the globe into quarantine—and many quarantined people into virtual worlds, hungry to connect with others. Second Life saw a spike of old, dormant users. Ebbe ran Linden Lab when growing awareness of virtual worlds fed into renewed and rising interest in the Metaverse.

Ebbe Altberg's passing was announced on June 4, 2021.

Ebbe did not quite live long enough to see Second Life return to public awareness. But a month after Altberg's death, Mark Zuckerberg announced his intentions to turn Facebook into “a metaverse company.”

In a keen irony, Facebook's name change came almost exactly 15 years to that day at the Commonwealth Club, when Second Life's Metaverse vision was hailed while a very young Zuckerberg sat to the side, almost like an afterthought.

Now Zuckerberg was being acclaimed as the Metaverse visionary, while Second Life was being remembered as one of Meta's ancient predecessors.

Now it's for us the living to ponder how this outcome might have been different.

Part 2: Postmortem

A couple key factors contributed to Second Life's failure to gain mass market.

Through no one's fault, for instance, Linden Lab had architected itself into a technical corner. In 2003, it seemed completely sensible to create a program optimized to run on a desktop computer with dedicated broadband. The succeeding years relentlessly chipped away at that assumption, as consumers shifted their overwhelming preference to Wi-Fi connectivity, and then to laptops with limited 3D graphics capabilities. And then starting in 2007, stampeded to the iPhone and later Android. In fewer than five years, Second Life seemed conceived for an obsolescent era.

“We missed mobile,” Philip Rosedale openly acknowledges now. “We paid a high price for that. Just like Facebook, by the way.”

Yes, around that time, Facebook was also slow to create a native mobile app of its own—but then rapidly changed course. (As we'll see in the next chapter, these moves are intertwined.)

“It was too soon,” Charity Majors observes, offering a postmortem from an engineer's perspective. “Both on the server side and the client side. The technology needed in order to make the experience real just didn't exist yet. Super basic fundamental stuff you need in order to scale a platform, we had to build or invent or hack away from scratch. Which is a testament to just how incredible the caliber of engineers was that we had there at the time.”

But the technical issues per se did not undo Second Life's mass growth, for other, similar platforms launched in the same era have gone on to grow much larger.

Hunter Walk, reflecting on his time at Linden Lab from the vantage of his subsequent years at Google and now as a leading venture capitalist, muses around similar counterfactuals:

“A different, simpler economy might have created more incentives for the audience to grow,” he tells me. “I guess I would have loved to give it a go where we could have designed around, ‘You need a certain degree of experience in terms of frame rate and concurrency’ and then said ‘What's the creation model and simulation model that allows us to maintain that density and simplicity?’ More LEGO-like but still interesting.”

Or, if I might rephrase that another way: What if Second Life had been more like Minecraft and less like a second life?

End Game

“Second Life not launching as a game first was a mistake, if the end goal was the largest audience possible,” Richard Nelson puts it now, though he notes that that direction might have come at a serious cost, at least in the beginning:

“I think if Second Life launched as a game first, we wouldn't have had the fun wacky crowd we did get [early on] that was into the metaverse concept.”

Having finally left Linden Lab in 2020, Nelson makes that point from a unique perspective: as a technical director at Roblox. (More on that in Chapter 4.)

Charity Majors agrees that Second Life should have had more game-like experiences at the start but believes the company's culture would have rejected them:

“All of the things that would make getting started catchier, faster, and more fun were things that I feel like leadership was violently against because it would have violated the vision.”

Core to that vision was keeping Second Life an open creative space.

“[I]f your world really is a game, does it overwhelm everything else going on?” as Cory Ondrejka frames the challenge now. “And is that right? And with Second Life, our choice, right or wrong, was no, we don't want a singular drive to overwhelm the creativity in Second Life.

“Maybe the right answer lies somewhere in between, where Second Life really could have used things like a stronger link mechanism, a stronger sort of textual overlay; we could have done more powerful search style structures, which then leads itself to sort of other styles of leaderboards that aren't as easily gamed, and that aren't so singular in how they drive behavior.”

In other words, content and activity ranking systems that SLers could enjoy as a kind of competitive mechanic, if they so choose.

I put the question to Philip Rosedale in the form of a hypothetical.

“If you could go back and make Second Life more like Minecraft, Roblox, and Fortnite put together, would you have done that? If you knew you'd end up with a very, very large user base—mostly of kids, perhaps, but also a much larger user base of adults?”

Philip doesn't quite answer in the affirmative but doesn't quite disagree:

“I think that looking back, if we had been even more focused on building tools that enabled gaming-type experiences, there might have been a very substantial lift from that,” he tells me.

“My passion was always for connecting with people. And so I think that it was always important that we do that. So I wouldn't have been happy, I probably wouldn't have been willing to work on Second Life if it supported games to the exclusion of supporting expression.

“I would have wanted to stay with a lot of rich self-expression and communication, but maybe we could have done things that enabled game-like experiences in a way that would have both been satisfying to a larger audience and maybe even been a better on-ramp. So I guess I agree with you in that regard.”

This cultural resistance to Second Life not being a game had real consequences that not only hurt user growth but caused Linden Lab to ignore easy opportunities—often to the irritation of the community creators who would have benefited from a growing consumer base.

Gizem Mishi of Blueberry vented about this in a phone call to Ebbe Altberg.

“I was very frustrated that I'm putting in all this work, and I feel like I hit a ceiling,” as she put it to Ebbe. “How can you do this to us? We deserve to grow!”

She demanded to know, for instance, why Second Life was not available on Steam, the world's largest distribution market for PC/Mac/VR games. The Sims franchise was available on Steam; Garry's Mod, a popular UGC-driven sandbox game, became a hit on Steam, selling over 20 million copies.

But as Gizem tells it, Second Life on Steam was simply not an option Ebbe would even consider on a conceptual level.

“I am not joking, his answer word-for-word was, ‘Second Life is not a game.’ I will never forget him saying that.”

The misconception persists today, even beyond Second Life; it is fashionable among many tech evangelists to argue that while the Metaverse may begin with game platforms, it will inevitably slough off its ludic carapace to become an immersive Internet with many multiple-use cases, completely separate from gaming.

In my view, all this misses a fundamental realization:

The very act of pretending that 3D graphics are a “world” inhabited by fantastic avatars is itself a game—as primal and basic a rule set as what we formulate on the fly while playing with each other on the playground as kids.

The grass is lava. Dinosaurs are here, and they want to eat us. The pixels on the display screen in front of us are actually a humanoid fox who's also an officer in Starfleet Command.

Fixated on turning Second Life into a “3D web,” Linden Lab lost sight of how virtual worlds and the Metaverse itself are and must always be essentially whimsical and playful.

What's in a Name

Another core friction point was not technical or related to the UX but perhaps the most subtle and the most profound of all: calling it “Second Life” in the first place.

The virtual world's naming, led by early Linden Lab business developer Hunter Walk, seemed to make sense at the time. As I wrote in The Making of Second Life:

“A lot of the game worlds were verbs that described what you'd be doing,” as [Walk] recalls it now. “You know … ever-questing! I wanted the name to be a vessel that people could fill themselves, that would be evocative of the promise of the world, and then put that responsibility upon the user to fill the promise.

“It's such a strong notion, it's such an idea: everybody wants a second life. You see more and more people with avatars and screen names,” he told [Linden management]. “I think if we hit our stride we can sell this. It won't be geeky. It'll be ‘Of course, why wouldn't you want one?’ Because it doesn't have to be better or worse, it doesn't mean your first one is lame or great, it just means it's different, and you can be somebody different, and do something different.”

His original suggestion, “Life 2,” was then massaged by the team into “Second Life.”

In the fullness of time, we can confirm that consumers do generally prefer metaverse platforms named after verbs, or at least strongly imply them: Minecrafting, building blocks and forts in Roblox and Fortnite, VRChat-ing, and so on.

Looking back at it now, was it a mistake to name the platform Second Life?

“I don't know,” Hunter Walk answers after a long pause. “It was a profitable startup, but it obviously did not fulfill the reach and potential that I hoped for. It feels right to ask in retrospect: Did that seem too geeky, describing what it was as opposed to what we needed it to be? I'm not sure a name with a different product would make a difference….”

From my vantage point, calling it Second Life created a double blade of skewed expectations, alienating most potential users (especially gamers) while also attracting a group of people who did very specifically want a second life—which alienated the first group even further.

“I think that's totally true,” Philip Rosedale says, when I ran my analysis by him. “The very idea that it was a second life and that it was separate from your first life and maybe that reflected mimetically or negatively on your real life.

“I didn't think about it at the time, because in my mind, I was only thinking about the Second Life,” Rosedale says, emphasizing the first word. “I think we also thought it was going to be a temporary problem.”

Adds Walk: “Maybe the way some people reacted to the name reflected on what we were prioritizing in terms of what the world would be like, and led us down the path of making a microverse and not a metaverse.”

This is also true: To the extent that the name was embraced by the dedicated user base, many or most of them interpreted it to mean a kind of roleplay luxe life they want but cannot have in the offline world—which is partly why the virtual world now teems with virtual seaside mansions, high-end shopping malls, and sexy nightclubs. (This direction was also shaped by the choice of default avatars, but more on that in Chapter 9.)

Paradoxically, while the name Second Life did contribute to the virtual world's lack of growth, it also helped attract excessive media coverage. With their game-like monikers, no competing metaverse platforms, even those far larger in terms of users, have received anywhere near the same level of outside attention.

Without that name, observes Philip Rosedale, “[H]ow else could we have done it, and gotten the kind of fascination that we did get? I don't know that if I find a time machine [to change it], I'd hit that button. Because I don't know what other name I'd use.”

Choose Your Own Inertia

Philip Rosedale created Second Life to connect people together, to enable them to communicate in a genuine way with anyone. But in the end, it may have been the breaking of in-person human connections that hobbled that vision.

In the virtual world, online communities generally come together for the sake of mutual fun and creativity; they evolve together around those goals.

In the offline realm, companies function from a panoply of interests, especially in the context of the Silicon Valley startup, where some staff are motivated by the creative challenge or even idealism, others by the chance of sudden wealth and fame. (Most, probably a mixture of those motivations.) And while California startups may boast more amenities and quirkiness than the average U.S. company, the corporate hierarchy and governance structure is generally the same.

The Tao of Linden was an ambitious bid to transform the corporate hierarchy; it sought to offer Linden Lab employees a culture that was about as creative and freeform as the virtual world they were creating. Its success, however, was decidedly limited.

“[T]he whole ‘Choose your own work’ went way too far,” Charity Majors says now. “We had no real management. And I get that; I hate managers too. But I also feel like it contributed to a lot of our floundering and our lack of direction.”

Charity believes this lack of management ultimately hurt Second Life as a product:

“Profoundly. Profoundly. We were not executing at the level that we should have. And we didn't have the kind of clear direction that we should have. There was a lot of cultiness at Linden Lab. We kind of told ourselves a story about how the rest of the world wasn't as special or as fun or as interesting or whatever.”

Engineer Richard Nelson concurs: “[I]t led us to adding all sorts of potentially interesting, yet incomplete features into the product, with not enough consideration for polishing the existing feature set or building what the users really needed first. I was certainly guilty of this.”

Any Second Life user should feel a glow of recognition at Richard's words: To this day the viewer software is crusted over with features and options that resemble an MMO game welded to a 3D editor duct taped to a social network crammed into an ancient television remote with hundreds of buttons.

“I think Linden learned their lesson eventually,” Nelson observes, “but the reputational damage from having a fussy, overcomplicated experience for new users had already been done by that point.”

Asked about that now, Philip Rosedale acknowledges the “choose your own work” principle in the Tao of Linden had a limiting effect.

“I think that, in retrospect, is not the way everybody should work,” as Philip puts it. “I was imagining that everybody in our company would have the same personality, and the same way they wanted to approach their work was just comically not true.”

Many if not most people at Linden Lab did not want the pressure of choosing their own projects.

“If I come to you and you can't justify what you're working on,” as he puts it, “I'm gonna fire you. That's pretty Elon Musk, right?” (We were chatting around the time that Musk had taken the reigns at Twitter, immediately going on a firing rampage.) “And that, I think, is an example of where I was wrong, because it was too much to ask somebody to gather all the information and then make a high quality, highly defensible decision about everything they're doing.

“As an adult now, as a grown-up, I recognize more. My ideas were in many cases neither unique nor correct. I would say at Linden, at least, we did some interesting experiments.”

Ondrejka blames disagreements with Philip around the “choose your own work” principle as leading to his ultimate defenestration from the company.

Looking back at it now, Cory believes if he and Philip were wiser back then, they could have compromised around their management differences.

“And instead, we just kept disagreeing and that's just super unhealthy, and it wasn't good for either of us, it wasn't good for Linden.”

At any rate, Philip Rosedale doesn't agree that he fired Cory Ondrejka over differences in management direction.

“No, no, I don't remember that. I mean, first off, Cory's great, I've seen him over the years.

I got into conflict with him about how he and we—both conversations are valid—disagreed about how he did things, how we did things.”

Philip doesn't get into specifics, but he describes it as ultimately a lack of wisdom on his part, with Second Life and Rosedale himself hitting peak media attention when he was in his mid-30s.

“I think in retrospect, I think I probably should have tried harder to understand, to listen and understand where Cory was coming from. And instead, as a younger man, I was just kind of not wanting to argue too much, and I was avoiding tough conversations. There was no one thing with Cory and I; there was no one smoking gun or something.”

In any case, what is indisputably true is that Cory Ondrejka's departure set off a chain reaction at the worst possible time.

“When I left, and when Cory left, and when others left after, that definitely had an impact,” Philip Rosedale tells me now. “Because we were the original ones that were really living that dream so much.”

“[S]oon after Philip steps aside as CEO, that kicks off five to six years of just a ton of turnover, both really seasoned employees and managers, senior leadership,” says Cory. “And all of that happens at a time of the world changing technologically. And having those two things hit at the same time, particularly with a product that fundamentally hadn't really found a true fit yet.

“I think that that really put Linden in a challenging position as a company and Second Life in a challenging position as a product … its evolution slowed down, just to a crawl.”

Second Life still does not, as of early 2023, have an official native mobile app. The company could have created one with minimal features early on, but it would be a monumental task to create an app compatible with the core experience and retain Second Life's essence; it would demand an epochal shift in resources and corporate focus.

It would have required, in other words, the cofounders unified around that direction and leading this charge. (Assuming that the Tao of Linden had been changed to “Choose your own work—as long as it's mobile-first.”)

Ondrejka also blames the rift between him and Philip on his own relative youth back then.

“I think both of us would say: I wish we’d the tools 15 years ago, the maturity, to figure out where we were willing to disagree and keep rolling forward.”

Cory credits that realization to a lunch meeting he had with Philip in 2019, at Orchard City Kitchen near San Jose. They had reconnected and briefly met from time to time since 2009. But it took a decade, by his lights, “when we both had enough time to reflect.”

“That was the first time I think we were both able to really see the magic we had together. Second Life doesn't get built by either of us alone.”

To be sure, Cory allows now, he and Philip still might have failed to push the virtual world into the mass market.

“[But] I think Second Life would have had a better chance if we were still there.”

By December 2008, within a year after both Ondrejka and Rosedale had departed, the passage in the Tao of Linden reading “At Linden Lab, you are expected to choose your own work” was quietly excised from the company wiki.

It was instead supplanted with a decidedly vague adage: “Good people make good choices … and vice versa.”

As for Cory Ondrejka, his involvement in the Metaverse was far from over then. But that's a story for another chapter.

As for me, I should end this postmortem by bitterly gnawing on my own share of the blame for Second Life's failure to reach the mainstream. Distracted with writing my first book deal, made over-giddy by constant media requests to explain this strange, exciting virtual world, I didn't do enough to manage expectations and advocate for the changes I've unfolded above.

During the apex of Second Life hype, the platform needed a far harsher critic who could point out the many shortcomings that needed to be resolved before the virtual world was anywhere near ready for prime time. I made little attempt to become that, at least not before it was too late. Fixated on the amazing avatar stories I kept stumbling into, airing technical and usability complaints seemed trivial, an issue to be addressed by others.

And when Linden Lab management announced that Second Life would no longer be called a game, I convinced myself that Second Life was becoming, after all, “the 3D web.” (Some passages from The Making of Second Life reflect that positioning.) For years, I'd hear dedicated users confidentially declare, “Second Life is not a game,” and for quite a while, didn't have the heart to gently point out that they were mainly just echoing corporate PR messaging.

Or to summarize all this in 50 words:

Second Life failed to go mainstream because its developers didn't realize in time that a metaverse platform must first succeed as an easy-to-play sandbox game, didn't transition to the social media and mobile era, and had an open-ended management structure that could not move the organization toward those goals.

In most cautionary tales of failed utopia, the perfect society falls to utter ruin.

But this is not one of those stories. For as Linden Lab the company kept struggling and failing to grasp mass market success, another mystery was still unfolding. This one is much happier.

Part 3: Second Life Forever

As I write this, Philip Rosedale remains a key adviser to Linden Lab and the evolution of Second Life as a platform. The company, for instance, is slowly rolling out its “Puppetry Project,” a new feature that enables a user to have their real movements and facial expressions captured through a webcam and other motion capture devices, then displayed on their avatar in real time.

“If you ask me what's a breakthrough area in the next five years where we could all go, ‘Wow, that's a huge, huge, huge improvement,’” Philip tells me, “I would say it would be animating the face and body of avatars with a [web] camera.” It might even attract a wave of new users interested in its creative possibilities for live performance and video streaming.

But the ultimate story of Second Life is being told by its virtual community, and it imparts a living vision of what the Metaverse can be at its best.

Against all likelihood, the established Second Life community keeps thriving within this glittering menagerie. An active and resilient community of some 600,000 active users, they held fast to each other and to the world long after tens of millions abandoned it.

As we'll see in succeeding chapters, the leading metaverse platforms in terms of population are, for the most part, dominated by minors, skewing heavily male, mostly people from the wealthiest nations of North America and the European Union.

By contrast, according to a 2018 demographic survey compiled by Nick Yee's Quantic Foundry, 52 percent of the Second Life user base is female, with a median age of 48.

According to 2019 data shared with me by analytics firm SimilarWeb, the virtual world was most popular for a time with people from Brazil, second only to the United States. And while we in the United States tend to take the Americanization of the broader Internet for granted, the rest of the globe may be heartened to hear that U.S. citizens in Second Life were just some 30 percent of the total.

These diverse demographics lead to genuinely inspiring virtual communities that rarely exist beyond Second Life's digital borders—certainly not in the social media realm, where activity is heavily siloed according to national origin, gender, and age.

In 2022, when Vladimir Putin's armored divisions rumbled into Ukraine, the country's cottage industry of metaverse creators was among the many overturned by infrastructure attacks that sundered them from the Internet. Among them was Hanna, the young creator of the beloved Second Life fashion brand OSMIA, who fled her city as Russia occupied it, even though she was six months pregnant.

The Second Life fashion community rallied to support Ukrainian creators like her with a purchasing campaign. And before she escaped to safe territory with her newborn child, a fellow creator kept Hannah's virtual world brand operational, so her family could still have that income to support themselves. Her friend is Russian.

Second Life is the only Internet platform I know of where it's quite common for people of different generations who are not related to enjoy each other's company, breaking through age boundaries by a shared interest in creative expression, conducted through avatars that rarely indicate real-life age.

The trans community is notably large in Second Life, with around 500 registered groups, comprising people from all ages around the world in search of a secure place to express their identity. One of them, named Kayla, encountered so much transphobia in her offline life that she expresses her authentic identity only through her avatar.

“When I found Second Life,” Kayla told my blog's columnist Cajsa Lilliehook, “not only could I be expressive of my gender, but also my real personality and visual styling.” Ishtar Angel, another trans community member, told Cajsa a similar story: “It was Second Life that actually helped me realize I was trans. I may be stalled in my transition, but I would not give up knowing who I am and how I feel now to when I didn't know.”

As U.S. wars in Iraq and Afghanistan wound down, I started to notice that military veterans, separated by distance and social unease in real life, would informally meet together in Second Life to discuss their PTSD symptoms and other painful topics, and together start to heal. These informal avatar-based support meetings grew so rapidly and are so valuable, retired Marine Lt. Col. Jay Kopelman, director of a vet support organization, once said with wonder: “I know Marines who say that Second Life is working when nothing else has.”

Its very nature as a user-generated virtual world inspires seniors like Fran Serenade to create new communities, and feel such powerful affinity with her avatar that her Parkinson’s symptoms abate, inspiring academics to ponder how other seniors might also benefit from the Metaverse.

Second Life's music scene is varied enough that it can accommodate Skye Galaxy, a pop pianist in his 20s whose voice was so alluring, it brought Gizem into the virtual world, which subsequently changed her life—and also Mr. Charles Bristol, a bluesman in his 90s whose very presence as an avatar surprised me so much, I've never left myself.

Second Life's creation tools are robust enough to empower content creators like Gizem Mishi, who can build multimillion-dollar brands from Constantinople, and artists like Jeff Berg, who can craft dreamlike experiences with pixels, visited by thousands, and no matter where they are in the world, feel themselves being spoken to.

Gizem, by the way, is far from the only creator in Second Life to reach seven figures of income through their virtual content. In an informal survey I conducted in 2018, one in ten Second Life merchants surveyed stated they were making well over $100,000 a year from the virtual world economy.

Linden Lab recently confirmed to me that Second Life user creators had grossed a total of $86 million in 2021—roughly comparable (based on publicly known data) to what Linden Lab itself grosses annually from Second Life.

That last point should be stressed: Creators in the Second Life user community earn roughly as much in aggregate from the virtual world as its corporate owners. No other metaverse platform has reached anything like this level of equity with its own user base.

Jeff “AM Radio” Berg is far from the only metaverse artist in Second Life, for a burgeoning art scene remains, even boasting gallery events and installation unveilings, where the entire surrounding world is part of the expression.

With funding from the Canadian government, an artist known only by her avatar name, Bryn Oh, regularly opens the doors of her private island, and tens of thousands rush in, eager to see what latest whimsical or disturbing stories she is telling in dioramic 3D.

On the other side of the world, Japanese avant garde pianist Tia Rungray often performs before a live audience of avatars, while a visual collaborator accompanies him, one time with a flurry of abstract video images, another with two massive cubes rising up in the virtual desert.

If an implicit goal of the Metaverse is to be a microcosm of the real world and its people at their very best, as I believe it is, Second Life has succeeded at this. In the process, it has also become a microcosm of what the Metaverse can and should become on a far broader scale.

To summarize all that in 50 words:

Second Life continues to succeed as a niche virtual world and model for the Metaverse because its creativity tools are open and powerful, its user community is diverse, and its company shares virtual world revenue about equally with a user community whose members are free to create endless value and possibility.

While other, newer metaverse platforms boast far higher user numbers, almost none of them come anywhere close to meeting Second Life's high standards of creativity, equity, and diversity. But better platforms are possible.

Making a Metaverse that matters means making that happen.