--:-- Host: We absolutely do not need AR and VR in order to build the metaverse.
--:-- Host: I have a feeling I've heard the term metaverse. Well, that term is actually 30 years old, originally coined by Neil Stephenson in his book Snow Crash. Fast forward to 2022, and numerous companies are now building towards their version of the metaverse, including Neil himself, working on Lamina1, a blockchain company oriented around creators. And while the present metaverse doesn't perfectly mimic what Stephenson had envisioned, we had to bring him in to discuss the design decisions he's making, but also the evolving intersection between the metaverse and gaming, the involvement of AR and VR, the evolving role of IP, artificial intelligence. And of course, we had to ask him where he gets all of his ideas from.
--:-- Host: I hope you enjoy this episode. The content here is for informational purposes only, should not be taken as legal, business, tax, or investment advice, or be used to evaluate any investment or security, and is not directed at any investors or potential investors in any a16z fund. For more details, please see a16z.com slash disclosures.
--:-- Host: Hey, Neil, welcome to the podcast.
--:-- Neil Stephenson: It's great to be here. Thanks for inviting me.
--:-- Host: Well, it's really great to have you. And I'm excited to talk about the metaverse, probably the topic you're sick of talking about at this point.
--:-- Neil Stephenson: I've heard of it.
--:-- Host: You certainly have heard of it. You coined it around three decades ago. So we will definitely be diving into that. But before we do, I wanted to get your take on a tweet that I saw recently. I actually saw it last week. And the tweet says, "Science fiction novelist is the highest impact position in the tech industry." And we'll be, of course, crediting Francois Chollet, who said that. But what's your take on that? Any immediate reactions to that?
--:-- Neil Stephenson: It's an interesting take, and I think there might be something to it. I've been talking about this for a while, particularly with the Center for Science and the Imagination at Arizona State University. I was on a panel years ago with Michael Crow, who's the president of Arizona State, and he was kind of getting onto this theme. We were talking about the fact that in some cases, not all, but some cases, a science fiction novel can serve as a kind of template or a roadmap to organize the activities of a company or an open source group that's more effective than typical corporate communication strategies.
--:-- Neil Stephenson: So if you look at a big company where there's a lot of different engineering teams and creative teams trying to coordinate their strategy, there's a huge amount of energy and mindshare that gets burned on just trying to keep everybody straight with the same unifying vision, working toward the same plan. And typically that takes the form of PowerPoint decks and endless meetings and discussions. It seems that in some cases, it's possible for a science fiction novel to kind of replace all of that. If everyone reads the book, everyone kind of gets it. It acts kind of like a magnetic field that organizes all of the iron filings, so they're kind of aimed in the same direction.
--:-- Neil Stephenson: Of course, there's some science fiction novels where that's not going to happen because they're set in a very distant future, or they're just not applicable. And there's other ones that seem to have that effect in some cases. So as a result of that conversation, we actually set up a project to create a science fiction anthology called Hieroglyph that came out in, I think, 2013, 2014, that was meant to consist of stories that might have that kind of value or that utility to them.
--:-- Host: Yeah, because I think a lot of people look to nonfiction for insight into reality. But I think often we forget that we're storytelling creatures and that people within an organization, I like that you use this idea of a magnet, in order to get aligned, there almost needs to be a degree of emotion there to really foresee some sort of future. And I think what's been fascinating about many of your books is that they've stood the test of time. If we use Snow Crash as an example, three decades ago, many of the things that you predicted within that novel have come to be. Whereas many companies, because you brought up the idea that it's really hard to align people within companies, they struggle to set a mission or a vision for a couple years that people can get behind. So I think that longstanding alignment is really, really interesting.
--:-- Host: And I think one analogous thing that we're seeing in the industry is this idea of the metaverse. And many people are using this term. I think there's a degree of disalignment in terms of what the hell that actually means. I mean, to give a couple examples, are we in the metaverse as we're recording this conversation? If there's an online auction, is that in the metaverse? If I'm staring at my phone for an hour intently watching TikTok, am I in the metaverse? And so as someone who coined the term, and I know not everyone will agree with your definition or your perspective, but I am curious to hear from you. What do you think of as the metaverse per se?
--:-- Neil Stephenson: In general, just to address the specific things that you mentioned, I think anything that's being used a lot today probably is not very metaverse-like because in general when people talk about the metaverse they're talking about the next thing that's coming along. So if it exists today, that suggests that maybe it's a pre, largely a pre-metaverse thing. But there are some exceptions that we can talk about in certain kinds of games and other experiences that I think are kind of on-ramps.
--:-- Neil Stephenson: One of the most basic ones is, do we talk about the metaverse or metaverses, plural? My colleague, Tony Parisi, has got seven rules of the metaverse. And rule number one is that there's only the metaverse. There's not a bunch of metaverses. So that, for me, is a strong indicator. If I see someone talking about "our metaverse" or "a metaverse," I immediately begin to question whether they really got it. Whereas if somebody talks about "the thing, what we're building in the metaverse," that gives me more of a warm feeling about what they're doing. And I think that's because a central idea of the metaverse, at least in the book, is that there is just one of them. And that doesn't mean that it's all kind of the same thing everywhere. Far from it. It's an incredibly diverse range of experiences, but you can always get from one to the other by moving around in a single unitary space.
--:-- Host: Right. There's a degree of interoperability between one and another. I'm curious to hear your perspective on that, though, because I guess your point is that we haven't made it to the metaverse yet. But many of the on-ramps, you could say now, whether it's gaming or social or degrees of immersion that we have online don't really have these connections, right? So many of them are disparate from one another. And so do you think that will change? I know we're going to get into your company, Lamina1, and I know there's other efforts by parties in, quote unquote, the metaverse space. I think there was a consortium of 37 companies recently that came together to discuss this. But to get back to the question, do you see this changing? Because this has not been the precedent in the last couple of decades.
--:-- Neil Stephenson: I do see it changing. There's not going to be one top-down big boss that says, "Here's how it's all going to work." So the ad hoc arrangements of people trying to work together. I think that the closest things that we currently can see to metaverse-like experiences are games like Fortnite, Roblox, Minecraft, where you've got a bunch of people running around in a space. They're all experiencing the same space at the same time. I mean, there might be shards, but the idea is there that you're going to a place and you're sharing it with other people. They've got huge user bases and because of the revenue that those generate and just the sheer amount of experience that those companies can rack up by running those operations from day to day, they're kind of the leading candidates for coming together to build metaverse type experiences.
--:-- Neil Stephenson: And I know that they're all actively thinking about this and how to do it. And I suspect that they're talking to one another about how do we build a bridge from our game or our experience to your game so that a person could go from one to the other. It may take some years to really come together, but I think it will come together. And I'm excited to see it happen and maybe be part of it.
--:-- Host: I'm curious to know, though, if you think the incentives are aligned for that future to be the future that we move towards. Because if you think about it from the, let's say the game developer side, some of those game developers may have incentives to not want to interoperate, right? To keep their users within their game and not move to another game. And then you can also look at it from the consumer perspective that perhaps consumers or the gamers themselves, some of them may really want this interoperability, but some of them may actually want those developers to focus on advancing that centralized game that they love so much. And so I'm curious to know how you think about that. And also, I'm sure you've seen some of the pushback even within the gaming community around this idea of interoperability and the integration of Web3 or some of these open standards. And so what's your take on that?
--:-- Neil Stephenson: There's a kind of people talking past each other phenomenon where you've got game developers interacting with crypto bros on social media, and they don't always get along. That's putting it mildly, right? So the game developer mentality for their position may be like this: "I'm a game developer. I spent my whole career developing an advanced skill set that I'm very proud of. It's very important to me and building unified experiences for people to play in. And every aspect of the art direction, the sound design, gameplay, the backstory, the programming, the engineering, it's all just exquisitely tuned in to deliver a particular kind of experience." So when somebody shows up wanting to bring extraneous things into my game, it's bad on two levels. One is just that on an engineering level, it shows total ignorance of how games actually work. So it's kind of in the "not even wrong" category of bad ideas. And the other objection is aesthetic. So it's just kind of an abomination to think of bringing a lightsaber into Assassin's Creed or something like that. And it's an insult to game developers. And it's reasonable for them to feel insulted by those suggestions.
--:-- Neil Stephenson: Games like that may be a little bit like a Ren Faire. You go to a Renaissance Faire agreeing to step into a medieval world for a few hours because you dig that. You like being in that environment. And it's just not acceptable to have kind of modern distractions in that environment. On the other hand, there's other games that are popular, extremely popular, which I've already mentioned, Fortnite, Minecraft and so on. Those games are creative mashups by design. So you can go into Fortnite and you can see a four-person team that consists of Iron Man and John Wick and Obi-Wan Kenobi and the Silver Surfer, and they're all running around playing together, and it's fine. No one has a problem with that because it's just part of the vibe of those kinds of games. And that vibe, that kind of mashup vibe, is explicitly what's described in the book Snow Crash as how the metaverse kind of looks.
--:-- Neil Stephenson: There is a future for interoperability in the context of those mashup style games, and it's out of those kinds of games that we'll begin to see interoperability. That doesn't mean that it's going to be mandatory or that every game somehow is going to be compelled to allow extraneous garbage into their finely tuned beautiful worlds. And indeed, as you pointed out, the people who love those games and play those games would hate that. I play a lot of Valheim myself, and it's a beautifully designed game that's a coherent world. And it would be extremely irritating for somebody to show up in that world with a blaster and start vaporizing me and my Viking buddies. So I think we can have both. A lot of the controversy that we've been seeing the last year or so on social media arises from this friction between the sort of almost ideological commitment to crypto and interoperability that you see on the part of crypto bros clashing with designers who think it's stupid on an engineering level and kind of an abomination on an aesthetic level.
--:-- Host: Yeah, I think it's really important that you're bringing up the engineering side of this because it's not just how we want these worlds to look, but also how they align with our engineering capabilities. And of course, one way that these games are advancing is through augmented and virtual reality, or at least some of them are venturing into these new worlds. I'd love to hear your take on the importance of that, whether these metaverses do need to be in, quote unquote, 3D, or whether actually many of them can survive and continue to thrive in the two-dimensional world that many people are participating in. Because that really is the world that many of us are used to, right? Staring at our phones, staring at screens. And many people, even though they're not truly immersed, they're not in the third dimension, they feel quite immersed still. They feel like they stopped playing their game after five hours and they feel like they've truly been in another world. So what are your thoughts on the level of immersion required for this, quote unquote, metaverse?
--:-- Neil Stephenson: We absolutely do not need AR and VR in order to build the metaverse. And 30 years ago when I wrote the book, I had a different view of it. And so I assumed that it would be all about goggles. A lot has changed since then, and we've all learned a lot. Doom came out the year after Snow Crash was published. And it's kind of almost hard to remember a time when there weren't games like Doom, meaning games where your screen is a flat window into a three-dimensional world. And so if you had described Doom to me in 1992, said, "Well, you're looking at a flat panel screen in front of you on a monitor, but you're seeing a 3D world through it and you're running around in that world," I'm not sure if I would have understood it or believed that that could ever really work very well.
--:-- Neil Stephenson: But now, fast forward 30 years, the day-to-day world that we're living in is one in which billions of people routinely access three-dimensional spaces through rectangles on two-dimensional screens, be they the screen of a laptop or a phone that you're holding up in front of your face. And it works really well. And one of the really weird aspects of it is the primitive control scheme. So most people are using the WASD keys on their keyboard plus a mouse in order to navigate these worlds. Keyboards are a Victorian technology and yet the human brain is so adaptable that even as clumsy as that is and as antiquated as that is, WASD is a perfectly useful way of navigating around in 3D spaces.
--:-- Neil Stephenson: I'm going to talk about VR. AR is a whole different thing, but let me just talk about VR for a sec. Early VR, just because of the limitations on processing power and so on, had high latency and other kind of quality issues. And it was, I think, pretty widely believed, even as recently as maybe 10 years ago, that as latency got reduced, as the quality of the experience improved, that we'd see a decrease in the tendency of users to get motion sickness. And I think that there was a decrease, but it didn't go to zero. It went to maybe, the last I've heard is like maybe 5, 10% state-of-the-art quality VR are going to experience some symptoms. And in fact, I was playing a 2D video game just the other day where my friends and I turned on a new feature and we all had to stop because we were getting motion sickness. So imagine if you were trying to popularize television in the 1950s and you said, "We've got these great programs. We've got I Love Lucy. We've got the Ed Sullivan Show, entertainment for the whole family. Five to 10% of you are going to end up throwing up into a wastebasket after half an hour of watching this." Well, that's a really high bar to commercial acceptance of entertainment technology.
--:-- Neil Stephenson: For AR, it's just a different thing. I mean, by its nature, when you're in an AR experience, it is or it should be somehow tied to the environment you're sitting in. Because if it's not, it's just kind of bad VR. One of the most fascinating things I ever did was trying to make content at Magic Leap where everything that we built had to be aware of, in some sense, what was in the physical environment and be reactive. It's an incredible thing to work on. But because of that, I think it's kind of different from what most people talk about when they talk about the metaverse.
--:-- Host: I think so too. But I also wonder with both of these, there's significant engineering challenges. Obviously, they're different. VR introduces this idea of motion sickness, which I would definitely be part of that 5% to 10% that gets sick. I get sick in cars all the time. And so what I want to understand from you, though, is if we are able to solve these engineering challenges, so for example, if it becomes instead of 5% to 10%, 0.001% of people get sick while using a headset, are people really wanting to be that immersed if we can solve those engineering challenges?
--:-- Neil Stephenson: For me, if it's a good VR experience, I like going into that level of immersion for half an hour, 45 minutes, and having a really intense experience. Beyond that, just the experience of having this thing on my face, not being able to see my real environment, not being aware of what's going on, being kind of socially isolated from the people around me, adds up to something that for me is a relatively brief experience. Like I said, half an hour, 45 minutes, kind of like you might sit down and watch an episode of a TV show, then it's over and then you stop. For all day, or for gaming binges that go on for hours and hours, I don't think it's for me.
--:-- Host: It's kind of hard to predict because as we're talking about these engineering challenges, it is hard to imagine how we can layer on community or social prompts or how you actually engage with those people. Because to your point right now, if you go into VR, it is mostly an individual experience. And I think humans are naturally very social creatures. I want to ask you a question about, you can apply this to AR, to VR, or even something like a Zoom call. But something that I've noticed is that as people create these new virtual worlds, many different sorts of those virtual worlds, what they tend to do is apply the physical world and digitize it. So let me give you a couple examples. If people want to create some sort of social engagement structure for work, they create a digital happy hour, which is just a bunch of people in rectangles on a Zoom screen. Similarly, if they want to create real estate within, quote unquote, metaverse or some sort of gaming engine, it looks a lot like a house in real life, even though maybe you don't need a roof. There's no rain. There's no snow, right? Well, it depends what's coded into that world. But something that is important to note is that these digital worlds don't have or don't need to have the same structures as the physical world. And so do you think there's something that people are missing maybe in terms of the wonder, the opportunity, the ability to actually go without the constraint of the physical world, to get rid of gravity, to get rid of how large that world is, like a lack of scarcity? So do you have any thoughts there? Because something that I found very interesting about Snow Crash was that it was this vast digital world, but there were certain things. I think there was a limit on height that people could be, but mostly there was not the same level of restriction as we have in the physical world.
--:-- Neil Stephenson: Yeah, I mean, it's going to vary from experience to experience. I do think that almost any plausible metaverse that I can think of, there's going to have to be some understanding around the size of avatars. It just doesn't make sense and it doesn't work if you've got some avatars that are 100,000 times bigger than others. People will break that rule and come up with experiences where that rule doesn't apply, but if it's too far outside the bounds of a kind of unified experience and nobody knows what to expect, nobody understands what's going on, then I think you're going to see a rapid fall off of interest. You're going to see spaces that are appealing to a pretty small number of enthusiasts, but are less interesting or might even be actively off-putting to the mass audience.
--:-- Neil Stephenson: I mean, Tim O'Reilly has said that a good way to think of the metaverse is that it's a communications medium. Any communications medium is trying to reach the broadest possible audience. That's why TV is the way it is. That's why movies are the way they are. The things that you're talking about, like buildings that are realistic and have roofs and ceilings, you're totally right that we don't need those things in the metaverse. But it is what people are used to and makes people feel comfortable to be in a space where there's a roof over their heads. There are some evolutionary psychologists who've proposed that early humans evolved in a savanna environment. So it's got open space, open grassland where you can see for a long way, but it's got trees. They're not densely packed together, but there's sporadic trees that you can go and you'd be in the shade or climb up to get away from predators. That's an environment that we tend to reproduce in our built environment. I think people are comforted by shelter, by the feeling of privacy that you get when there's some walls around you, there's a roof over your head. So we're going to see those forms recapitulated in metaverse experiences that are geared towards drawing in a mass audience. But clearly, there's no limit on what you could build, what you can imagine if you're just trying to follow your own personal aesthetic or appeal to a kind of a smaller cult following.
--:-- Host: You have spoken to this idea of trying to attract the masses, which I think is true to some degree with consumer applications of gaming, AR, VR. But as you mentioned, you've worked at Magic Leap for quite some time, or you did work there. I think you were the chief futurist or something like that. Very cool title. What did you learn there? Because they seem to have pivoted now more towards commercial enterprise. What did you learn there about the potential applications outside of maybe the consumer lens and where some of these technologies can be applied to solve other problems?
--:-- Neil Stephenson: What I said at Magic Leap very early was that I didn't just want to be a guru, navel-gazing futurist, but I wanted to build something. I looked around for a little bit at what was going on inside the company and decided that probably the most useful place for me to apply what I know would be in content creation. So there were a number of projects there, and presumably still are, that are aimed at specific industrial commercial type applications. Just like you said, 99% of my time was spent building applications that were aimed at just sort of a general audience and meant to be entertaining, fun in some way, as opposed to the commercial, industrial stuff. You do have to solve a lot of the same problems, regardless of whether you're making a commercial product or an entertainment product. You're building an experience in a game engine. We use both Unity and Unreal at different points. For it to work, the game engine, the environment, the level has to be populated not just by imaginary entities that you drag into the scene, but by stuff that's actually in the user's environment that the system recognizes through machine vision and kind of creates in the world.
--:-- Neil Stephenson: So there's a loop that's running all the time where the cameras on the device are scanning, cameras and other sensors on the device are scanning the environment and trying to figure out what's what. Like, "Okay, there's a flat thing down there. It's probably the floor. There's a flat thing over there. It's probably a wall. There's a window. There's a door." And building up from that, you can make these systems more and more sophisticated and capable of recognizing more objects. But each one has to lead or should lead to a mesh, an asset being put into the level that your AIs or whatever are navigating around in. So that kind of base layer of functionality is to a large extent built into the operating system of that device. Building on top of that, we had to kind of create our own stack that started with engineering and sort of worked its way up to creative elements.
--:-- Host: A lot of the work that you did towards those creative endeavors for your project, how could you see those being applied to maybe some more commercial projects? I'm just trying to get the wheel spinning in terms of many people believing that these technologies are going to be applied to gaming, which is certainly true. And maybe even they might have a qualm with that idea with a lot of this technology only being applied to this idea of gaming. But I think there's a lot of potential, as we see with many technologies, for them to be applied in several ways and ways that even the original creators didn't expect. I think one example that I've heard you give is that electricity was obvious to be applied to light, but electricity then generating electric guitars, less obvious. So are there perhaps less obvious things that you've been exposed to or that you've started to think through, maybe even for a future book, in terms of some of these technologies being applied in ways that we might not expect?
--:-- Neil Stephenson: That's a great question. So I may have to think about it for a second. So for example, one of the two main projects that my team worked on at Magic Leap is called Baby Goats. And it simply populated your environment with baby goats that would run around, they would jump up on the furniture and kind of interact socially. But it wasn't a game, game. It wasn't like you're trying to score 10,000 goat points and kill the goat boss. It just was an ambient kind of experience of these creatures running around. And so it had to do all those things I mentioned. The goats had to know where the furniture was and they had to know, "Oh, well, okay, I can jump one meter in the air, so I can't jump directly to the top of the bookcase, but I could jump to the table, and then from there I could jump to the bookcase." So the furniture would be placed by the level designer, and the possible paths where the animal could move would be sort of pre-programmed. And so the animal's just making a decision to follow a certain pre-programmed path. And you can use some AI there too, if you want.
--:-- Neil Stephenson: In the AR version of this, you have no way of knowing what every person's living room is going to look like. And so you can't have any pre-programmed map of the room or plan. It's all got to be AIs kind of hatching their own plans on their own. You just push the code out into the world and hope it works. Typically you see emergent behaviors that you might not have expected. And we're seeing this a lot with DALL-E and Midjourney and other AI-based image programs where those things produce some images that are probably what the designers were envisioning or imagining. But there's also just crazy results that come out of these things that no one ever could have predicted. In the metaverse, I think we're going to see virtual environments that are populated and brought to life by AI. In some cases, it's going to be what you might think, which is non-player characters that walk around and deliver the mail or whatever. But I think we're also going to see whole other categories of AIs being put to work to create experiences that have a kind of realistic texture about them and that make the environment seem convincing to the user.
--:-- Host: I'm glad you brought up AI because it is such an important theme, at least within the zeitgeist currently. And another important theme as it relates to the metaverse is the idea of IP. Who owns a particular object? We talked about interoperability before, right? Can you move that between different engines? How do you think AI will interoperate with IP, if that makes sense? How will these AIs generate a bunch of objects, maybe even generate games themselves eventually? And then who owns those things? Obviously, the AI is trained on many different data sets. So do the people who supplied that data own the eventual creation from the AI? Or have you started to think about how that might work?
--:-- Host: My other co-founder at Lamina1, Peter Vessenes, has actually been thinking a lot about this. He went deep down the rabbit hole of Midjourney and all that last month. And very quickly, when you're thinking about applying those kinds of systems to build experiences that might make money, you're up against an ethical question for exactly the reason you said, which is that these AIs are harvesting work from millions of people.
--:-- Host: You might ask DALL-E to make a painting of a dog in the style of, and that's fine. But when you do the same thing and you specify a living artist, then you can be sure that the system is going around and ingesting data from everything that artist has ever created and then doing a mashup to produce a result. So why shouldn't—I mean, if you're just screwing around, it's one thing. In the terms of service of DALL-E, I don't know about the other programs, but they're pretty conscious of this and they're trying to prevent people from just monetizing the crap out of it. In cases where it does get monetized, why shouldn't the artists who contribute get a cut?
--:-- Host: We've been talking to Shrapnel. It's a game that's being put out by a company in Seattle called Neon Machine. And it's a blockchain Web3 kind of game from the ground up. So they've been doing some interesting work in the field of smart contracts that are structured in such a way as to track the provenance of a given game object and create what Jaron Lanier would call a value chain, which is a bunch of people who one way or another contribute their time and effort to making a thing. And if that thing ends up making money, generating revenue, then how do you reward them?
--:-- Host: In the motion picture business, they have this lovely term called a waterfall, which is in a contract for a movie. When revenue starts to happen, it cascades down and gets diverted off to different people. And there's a whole structure in place for this. That all works, but it's notoriously slow and cumbersome. So is there a way to replace all of that machinery to create a waterfall that basically just works through smart contracts and just automatically distributes a revenue stream among people who contributed to a particular project in varying proportions? And that is actually not a bad fit for crypto.
--:-- Host: Crypto has got smart contracts that can route money to wallets. Kind of the whole point of a blockchain is that you can announce that at a particular moment, this transaction happened. And everyone on that chain agrees that this is true. In the classical application of blockchain, you're just sending money to people. But in an IP-centric implementation, you could say, "I and Neil wrote a book on such and such date with a magic sword in it. And here's what the magic sword kind of looks like. It's about yay long and it's sharp on one edge." So that's an assertion that a creator could make that they had created this idea at a certain time. But it's a hell of a long way from being a game asset that can actually be used to create an experience. And as such, it's not capable of generating any revenue.
--:-- Host: Somebody could come along and then link off of that stub asset and say, "Okay, I'm going to make an implementation of this idea. I'm going to credit Neil for having the idea in the first place, but I'm going to add something on top of what he's already posted on the chain. And these assets turn this into something you could drag into an Unreal or Unity level." And then somebody else could come along and say, "Okay, I'm going to buy that from the asset store." But it's still a long way from working in my game because I need to set up blueprints or whatever. I need to change the art direction of it so it fits with my game. I need to do some programming. I need to change the sound effects. But it's still all part of the value chain that starts with the original asset.
--:-- Host: So what the people at Shrapnel are working on is some templates for smart contracts that would sort of ease the process of creating these chains and organizing them in such a way that the waterfall of revenue would happen if any of it ever actually managed to make any money.
--:-- Neil Stephenson: I think maybe if people are listening, they might think that the description that you just gave sounds very far away. But actually, it reminds me of something I saw this morning on Twitter. One of the other AI engines—there's DALL-E, there's Midjourney, as you said—another one is called Stable Diffusion. And I saw this thread this morning of a bunch of Redditors who were using clip art or very, very simple drawings. You just imagine like a circle. And then they were using Stable Diffusion to basically take those and turn them into assets that looked like they were from a game, really crisp, really high definition images of these creatures or participants in a game potentially.
--:-- Neil Stephenson: And so this is the future, right? Using these different engines to upgrade prior people's contributions. And I think you're right that you need to somehow articulate that value chain because we're used to items in our homes. Imagine the physical equivalent. If you had a basketball and you said, "I have this machine that turns this basketball into a car," that would be wild. But you actually can do some of this stuff in the digital world with some of the tools that are advancing. But you do need to know the value machine, as you said, who created what and how did that lead to the next creation which led to the next creation and so on.
--:-- Neil Stephenson: You also want a feedback loop that runs in the opposite direction. Somebody might come along in my example and create an implementation of my sword idea that I just hated. Either they did a really shitty job of it, or maybe they're just a troll who is deliberately trying to be offensive. And so in that event, I should have some kind of right of approval. I have to have that. You can imagine situations like if I created an avatar that looked childlike, it would be fine for somebody to take that and use it in a children's experience. But in certain other kinds of experiences that we can all imagine, sadly, it could be a very bad abuse of my creative work.
--:-- Neil Stephenson: And so if I created such an avatar, I need to have the ability to say, "You can use my work in this thing that you're building." And because it's too much work to examine every possible site where something like that might be used, you then need to go up one level and have kind of rating systems, or third-party evaluators that look at sites, look at experiences and say, "I don't think you want your IP to appear on this particular site or to be used in this way."
--:-- Host: That's actually fascinating to imagine because you do see the parallel of this with 2D work. A lot of these AIs are being trained on the whole of the internet. But even prior to the AIs existing, you had many photographers, as an example, who in some cases were very protective over their work and had ownership of it and could determine where they wanted it or where they didn't want it. And then you also had the emergence of things like Unsplash, where you had creators who said, "Actually, I'm okay with this being utilized by anyone, anywhere." And you'll probably see the same phenomena happen with what we're talking about.
--:-- Host: But in the case where people didn't want their work utilized in places that they hadn't approved of, there were these third parties. There are sites that actually go out there and scrape the internet to find images that are owned by people that have copyrighted them. And then they go back to those individuals and say, "Hey, did you know that your work is being used on X, Y, and Z places?" They're a little bit like extortionists in some ways because they hold the people who use them incorrectly accountable. But it's fascinating to imagine that some similar approach will be needed in this world of AI as well.
--:-- Neil Stephenson: And there's many other examples. I mean, I've picked one that's kind of disturbing, but there's just routine ones. Let's say that I create a piece of visual art and sell it to somebody in the expectation they're going to hang it on their wall of their virtual home or their real home or just display it for their friends and family. Then what if they turn around and use it as the logo for a new brand of basketball shoe? They're using it in a way that I didn't anticipate when we set up the original deal and it's all over the world. And maybe I should get paid for that. Or maybe I should have the ability to say, "I don't want my art used in that particular way."
--:-- Neil Stephenson: So there's a whole world of IP rights and licenses that right now is not being handled at all by the way NFT marketplaces are currently set up. And I think the next wave of NFTs is going to begin to feature some of these improvements. Or I think they're improvements. Some people are going to hate this.
--:-- Host: Well, actually, I like that you brought up that you think they're improvements. Other people may not think so. And I think technology is often viewed that way. Some people believe it's very utopian, some people believe it's dystopian. There's probably a huge spectrum where any given individual technology can actually be applied across the spectrum in positive and negative ways.
--:-- Host: And I think actually, even using your books as an example, *Snow Crash* is more of a crudely rendered dystopian metaverse, if we're going to use that term. Another one of your books, *The Diamond Age*, has a more utopian vibe. It's more organic, more communal, more people building towards something more consensus based. I'm curious to hear from you—and feel free to loop in your new project, Lamina1—how do you think about building technology so that it is leaning more utopian? Or maybe utopian is not the right word here, but how do you design technology so that it can be used for good? Or is it just a reality that humans will use all technologies for good and evil, and that's just something that we can't avoid?
--:-- Neil Stephenson: I think that is pretty much a reality. And people are endlessly creative in both good and bad ways in how they use things. *Snow Crash* is absolutely a dystopian novel. It's also kind of making fun of dystopian novel tropes in a lot of ways. It's both a dystopian novel and a parody of dystopian novels. The technologies that are shown in it, with some exceptions, don't really have any bias for bad or good. A lot of what we see in the metaverse in the book is kind of garish and crude, but that's just what we happen to see.
--:-- Neil Stephenson: It's also shown in the book that Hiro has got a beautiful home in the metaverse, kind of Japanese style home that is exquisitely rendered. And it's a fine work of art. And the character of Ng has also got a beautiful home of his own. So another example would be Earth. There's an application called Earth in *Snow Crash* that has a lot in common with Google Earth. It'd be silly to say that because Earth was described in a dystopian novel, that it's an inherently dark or dystopian kind of application. When we look at Google Earth today, I think the vast majority of its applications and overlays are for interesting, creative purposes. And there's a few ways in which it can be abused. But we see Google managing it fairly well in the sense that they go out of the way to blur out faces and try to prevent darker or less constructive uses of the technology.
--:-- Neil Stephenson: There's not like built-in biases towards bad or good. But I do think that one really important technological bias that is kind of baked into current social media has to do with the revenue model. I don't think that this was evil people coming up with an evil scheme. I think it was just engineers saying, "Okay, let's give people more of what they react to." So as we all know, social media platforms tend to have built-in feedback loops that feed people more and more emotionally powerful content, regardless of whether that is constructive. And the reason they have to do that is because that is their revenue model. I think now that we've learned that, maybe we can avoid recreating that in the next wave of metaverse development by thinking about how these experiences are going to make money. What I'm hoping is that we'll end up with a revenue model that's more obvious and transparent.
--:-- Host: Do you think that people will opt into that revenue model? And I ask this because I think you're right that social can be hindered by its revenue model, by its nature. I'm sure you've heard the saying, "If you're not paying for a product, you are the product." And so I think maybe the awareness of that phenomenon will increase over time. And maybe some people will opt into other revenue models. But there's a reason it works. Many people gravitate towards these large social platforms because that's where you can have a platform. That's where you can reach so many people. And the masses, for the most part, have okayed this idea where they are giving up certain privacy rights and certain aspects of their information in order to not have to pay for things. So do you actually see that reversing? Is it an awareness thing? Or how do you actually imagine a revenue model that you can get the masses to opt into that is not an ad-based model?
--:-- Neil Stephenson: It's a really tough thing to turn around just because the attraction of free stuff is so powerful. There's an analogous phenomenon in games, casual mobile games, where at a certain point it just became the expectation that games would be free. Sometimes that means you have to watch ads, which is annoying, but it's a transparent kind of annoyance. You understand how the wheels of commerce are turning underneath that experience. But in other cases, these games become like slot machines where they're using applied behavioral psychology to make people want to keep punching in their credit card number and spending money. The attraction of free stuff is incredibly powerful and it can lead to all kinds of negative social effects.
--:-- Neil Stephenson: I wish I had an answer. The person who I think has done the most useful thinking about this has been Jaron Lanier. And we talk about this all the time. He's got ideas for structures that would be analogous to labor unions where—right now the way social media works is this: if a bunch of people all volunteered their time to build a car and then the car company sold the car and made money off of it, which is pretty amazing that people are willing to do this. The idea that Jaron's working on is one in which you would join up with a bunch of other people with like-minded similar interests through collective bargaining, essentially sell your data to a social media company. There's ideas out there, but beating the draw of the power of free stuff is incredibly difficult.
--:-- Host: Yes, I think we can all empathize with that. I can't tell you the number of times I've gotten an email that something was free and I've wasted my time on it. And the thing that I got free was something I never even wanted. So for things that you do want—connection, community—that you get through these applications, it's no wonder that that is a very, very tough phenomenon to adjust for so many people.
--:-- Host: That brings us to your new company, Lamina1. So you are actually participating in trying to build the infrastructure that may change some of these dynamics. From your website, it says: "Lamina1 is a new attempt at the base layer for the open metaverse that privileges creators, technical and artistic, one that provides support, spatial computing, and a community to support those who are building the open metaverse." Can you share a little bit more about why you're choosing to build Lamina1 and why you've made some of those really concrete design decisions like building on the blockchain?
--:-- Neil Stephenson: I think I've already hinted at a lot of this in the earlier parts of this interview. So a lot of what I've been talking about is all kind of rolled up into Lamina1. For those who don't track crypto stuff, a layer one blockchain just means a new blockchain. It can be built using new code, new technology. In other cases, it could just be a fork of an existing chain or a combination of both. But the idea is that when you use an existing chain and build your stuff on top of it, you're essentially going along for the ride with whoever runs that chain. You don't have a say in how it's organized and what engineering characteristics it has. But if you start your own layer one chain, you get to control that. You get to build in the features that you want for your particular application.
--:-- Neil Stephenson: Bitcoin was the first layer one chain. Ethereum came along later. It's a separate layer one chain, but it adds features that weren't present in Bitcoin. We think that the next wave is going to be metaverse applications. We think that cryptocurrency and fiat currency—old fashioned non-crypto currency—they're both going to be used in the metaverse just as they are in the real world because they both work and they're so good for different things. To the extent that crypto is useful in the metaverse, there's different layers to it. Money is just the most basic layer. So having tokens that act as money and that you can send back and forth between people to pay for things is the most basic functionality you can have in a chain. Smart contracts are built on top of that. And there's various knobs that you can turn to optimize a chain to carry out specific functions.
--:-- Neil Stephenson: We think that for some of the reasons I was mentioning, having to do with how content creators can get paid to build experiences, there's an opportunity to make a new layer one chain that works for them and that's integrated with the tool sets that they are in the habit of using. So they don't have to go out and become crypto experts and learn all kinds of new capabilities just to do their jobs. Making things good for those experienced creators is the key to having a successful metaverse because nobody's going to go and use the metaverse unless there are experiences there that they enjoy having. People who know how to make those experiences right now, by and large, work in the game industry and they're good at using game engines and the tool chains that feed assets into those game engines. If we can find ways to integrate the financial infrastructure that they need to get paid into the tool chains that they're already using, then we like to think that that'll bring in content creators, make them happy, create experiences in the metaverse that lots of people enjoy having and are willing to pay for. That's kind of the plan in a nutshell.
--:-- Host: That makes sense to me. And I think, whether you're in the metaverse or otherwise, every company out there is trying to attract top talent. So if you can acquiesce them towards your ecosystem, then you likely will win. But when it comes to the incentives to drive creators to participate using Lamina1 or otherwise, how do you think about that? Many of these ecosystems will say, "We'll give creators this percent cut in order to develop within our metaverse or another metaverse." How does Lamina1 layer into that? How aware is the creator, the end creator? How do they know whether they're using Lamina1 or the gaming engines that you mentioned? How do those all interplay together? Is Lamina1 almost like a transparent layer under these existing engines? Or can you share a little bit more about how that might work?
--:-- Neil Stephenson: I'm not sure if I would say transparent because that suggests invisible. And so to your point, people should know what they're using and why they're using it, how they're getting paid. Anyone who is trying to make a buck and support themselves financially has got to be mindful of where the money is coming from and how the business operates. A lot of times the interfaces that you have to use in order to do anything at all with crypto, even pretty simple things, are hard to work with and confusing and almost a little frightening because you're sort of aware at some level that if you do it wrong, you could lose money. Your wallet gets wiped, etc. We're certainly fans of making it easy and accessible without having to learn a bunch of new stuff.
--:-- Neil Stephenson: I think it would be a mistake to have it be so transparent and so invisible that people don't even know what's happening. That needs to be very frank and obvious and upfront to creators. And I don't think that creators would accept anything other than that, because again, this is how they're going to get paid. This is how they're going to make money. And so it needs to be out there where they can see it.
--:-- Host: How are you forging these partnerships with the gaming engines as well? Because I assume that if these creators are building on top of existing ecosystems and you're building a protocol that in theory interoperates with them, how are you actually building those relationships? Or are these gaming engines super open to integrating with Lamina1? I'm less familiar whether these APIs are already open. But tell me a little bit more about that.
--:-- Neil Stephenson: The two big ones are Unity and Unreal. And people make plugins for those all the time. Unreal is available source. It's not open source—that's got a particular legal meaning—but you can download the source and compile from source anytime you want, which means you can add whatever you want. And I'm less familiar with Unity, but they've all got scripting capabilities and ways that you can add on to the engine.
--:-- Neil Stephenson: At the moment, most of the big players in the game industry are being avoidant of crypto and blockchain. And there's different reasons for that, having to do with they may perceive a risk to their brand, or they may see that there could be a lot of headaches administratively trying to change their business model to one in which lots of real money is being handled by their customers. So at the moment, they're taking a hands-off approach. In some cases, it's pretty in your face, like Minecraft made a pretty strong statement just a couple of weeks ago about not wanting to have crypto stuff going on in their platform.
--:-- Neil Stephenson: But at some level, when you talk to people in the game industry, whether it's individual developers or people working in larger companies, I think there's an awareness that this is coming and that it could be an important thing for their businesses and their companies in the future. If we can just get past the early kind of Wild West era, the initial teetering issues, and out of the prototype stage, and into a place that just feels more stable.
--:-- Host: And also once you start to partner with some of them, if you are able to attract that talent, the best creators, then it'll become potentially inevitable for the others to want to partner and open up to that ecosystem because they will be following the talent.
--:-- Neil Stephenson: Yeah, it could be. I think that people will happily continue using fiat currency and traditional payment schemes for a long time because they work.
--:-- Host: From where I sit, I just think that there are some specific things that we could do with a new chain that could be valuable for people in this space. And I know you've talked about them to some level already, but what are those specifics? Are there a few things that you think just don't exist within the existing infrastructure that Lamina1 is looking to fix?
--:-- Neil Stephenson: The big one is the notion of value chains. I went into some detail about that before, so I won't rehash it. But existing payment systems can be really cumbersome. If I hire someone in another country to make a skin for me in Roblox and pay them some money for that, does that mean I need to 1099 them? Do I need to fill out the paperwork in some other country to make sure that everything is properly squared away? In a 20th century or 19th century economy, it was okay if you're a large company, you just put up with that burden of paperwork. But in an internet economy where there's constant transactions happening all over the place, you want something that's a little more fluid. And I think cryptocurrency can definitely provide that.
--:-- Neil Stephenson: Likewise, I described the way work and waterfalls work in the entertainment industry. The Hollywood legal industry is something that you cannot believe until you've seen it. Just the size of it and the number of people who work there and the complexity of the structures that are created and managed. So if all you're doing is punching out a few blockbuster movies and handling the revenues from those, then that's a manageable situation. And it works. Lots of people are not happy with how it works, but it totally works. People are making money. Movies are getting made.
--:-- Neil Stephenson: But creating complicated experiences in a virtual world is not going to be like that. It's not all stovepiped into specific big budget projects. It's assets going every which way and appearing in different places. So trying to track all of that and make it work as a business using conventional Hollywood legal and accounting procedures just doesn't sound realistic to me. It won't work. Doing it with a new chain that's optimized to handle those kinds of transactions seems like a reasonable idea.
--:-- Host: Yeah, I think so too. I think another design decision that you've made with Lamina1, which seems to be adjacent to the things that you just discussed, is also to make the chain carbon neutral. Can you speak to why that design decision was made?
--:-- Neil Stephenson: We started talking about Lamina1 just a few months after I published a novel, *Termination Shock*, which is all about global climate change. So it would seem weird for me to publish that novel and then a few months later announce a chain project that's going to be putting carbon into the atmosphere.
--:-- Neil Stephenson: For people who are not crypto dorks, I should explain that old school Bitcoin uses an underlying system called proof of work, which relies on solving complicated math problems that inevitably consume a lot of energy. So it is terrible from an energy standpoint and from a carbon footprint standpoint. Newer chains, by and large, are going to a completely different system called proof of stake, which is orders of magnitude more efficient. So on a modern chain that uses proof of stake, if you're using this for routine transactions, the carbon footprint really is no worse than what you're already doing anyway. If you drive your car around or turn on your air conditioner or take a plane flight. In proof-of-stake systems, which Lamina1 will be, it's not that bad. But it doesn't matter. You're still at some level responsible for putting some carbon into the atmosphere.
--:-- Neil Stephenson: So what we're going to do is design the system. And this is one of those things that you have the power to do if you do a layer one chain as opposed to piggybacking on somebody else's chain. The system is going to be designed in such a way that in order to participate, to operate a node, you need to demonstrate that you've bought a certain number of carbon credits from a legitimate company. And that's a whole topic unto itself because there's a lot of different companies out there pushing a lot of different carbon credit schemes and some are more legit than others. So there's actually multiple ratings agencies now that all they do is look at these systems and try to figure out which ones are good. So it gets a little complicated on that level, but we're going to do that because it's the right thing to do.
--:-- Neil Stephenson: And we're pursuing more far out ideas related to carbon sequestration. They're too early to talk about, but it's on a separate organizational track from Lamina1 because it's still kind of in the gonzo, probably won't work phase.
--:-- Host: We're actually doing an episode on carbon removal. So very interested in this space and glad that you're pursuing it as well. I want to dovetail this into the final section, which I would be remiss not to ask you about, which is just how you think about the future. And we talked earlier about this idea of the metaverse being coined 30 years ago. Just to put that into perspective, that's actually before I was born. And so I'm very impressed as I read *Snow Crash* just to even get my mind around this idea that you were able to come up with these terms, but more than that, to imagine a future where we are immersed. I mean, this is like when we had dial-up internet or even before that. And so that's impressive. But you've also done this across a myriad of your books.
--:-- Host: So you've been involved in the early stages of fintech, and we've talked about crypto. You've now written a book, *Termination Shock*, about carbon removal and climate change. This is a broad question, but where do you get these ideas from? How do you synthesize the seeds of these books when they're so far ahead of what many other people see in reality?
--:-- Neil Stephenson: A lot of times it's because I've had contact with the actual technology being described. In the case of *Snow Crash*, that emerged from a project I'd been working on for a couple of years in computer graphics, where I was writing a lot of code. I was trying to push hardware to do things it wasn't quite ready to do yet. That got me thinking about the future of 3D graphics as a medium. Having that direct involvement helps in two ways: it gives you ideas, and it exposes you to the specifics of how the system works, which produces ideas for plot devices.
--:-- In the case of *Termination Shock*, we have someone launching sulfur into the stratosphere for geoengineering. If you sit down and run the numbers on what it would take to do that, it's an interesting engineering problem. By thinking through the engineering details, you come up with ideas that wouldn't occur to you if you were just freestyling.
--:-- Host: I want to ask about one particular example. I've been lucky enough to work remotely for about seven years now, so I felt like I was ahead of my time. One of my favorite quotes from *Snow Crash* is Hiro basically saying that a particular video game job was managed by the Nipponese, which means all the programmers have to wear white T-shirts, show up at eight in the morning, sit in cubicles, and go to meetings. This was written in the early 90s as a small hint toward where the world was going. Where did that inspiration come from? What did you notice that many other people didn't?
--:-- Neil Stephenson: The 70s and 80s were a period during which the Japanese economic miracle happened. After the oil shock of 1974, people started buying smaller cars, the Japanese car industry went through the roof, and people became fascinated by their management approach. You'd see documentaries about factory workers showing up at a certain time and doing calisthenics before work. That was on people's minds in the 70s and 80s.
--:-- In my case, because of the project I described, I ended up working alone from home for a couple of years. I was a writer, but also the coding project was just me doing all of it. I wasn't the only person operating in that mode—this was still the era of the lone hacker who could single-handedly produce a gigantic system that worked. But you could see that era was coming to an end. We were heading toward a place where the individual lone hacker wouldn't be able to keep up with large organizations.
--:-- From a storytelling point of view, that gave me elements that were immediately powerful for defining my character. Hiro is a lone genius hacker who's really good at that. He sees himself in contrast to the regimented style of a certain kind of workplace that would never work for him. Yet he knows at some level that this is all going to change and that he's headed for the scrap heap of history. On his business card, Hiro identifies himself as the last of the freelance hackers. These are romantic ideas, tropes from literature that are known to be effective. Combining them in that particular way made it possible to do a new twist on an old story.
--:-- Host: Your answer was more complex than I expected. Even though the book was written in the 90s, you were looking decades before that at themes that had emerged leading up to it. It was written in the 80s, right?
--:-- Neil Stephenson: I started working on it in '88, '89, wrote it in 1990-ish. It takes a while for a book to go through the pipeline, so '92 is when it came out. It's very much a work of the late 80s.
--:-- Host: You just wrote *Termination Shock* and many other books. From the perspective of 2022, what clues are you looking at? What themes do you notice today that might be important for the next 30 years?
--:-- Neil Stephenson: To me, the two most important things are carbon and the polarization of society—the way polarization is being fostered and weaponized by bad actors who benefit from it. We've talked about both during this interview. I don't know what the answer is on polarization. A lot of smart people are thinking about it and writing great stuff. I was just talking to Anne Applebaum, who wrote *The Twilight of Democracy*. Anne and I met when we were helping Jonathan Rauch launch his book *The Constitution of Knowledge*, which is also about this. But what's missing is a solution. When you ask these people, "What are we going to do? How are we going to fix it?" no one has a silver bullet or anything close to it.
--:-- I'm thinking a lot about the other problem I mentioned—carbon. I know a lot about that at the moment, if I do say so myself, along with geoengineering. I rolled directly from a book about geoengineering into thinking about carbon capture and sequestration. That's an area where the science fiction writer mind can get more traction because you can think about it on an engineering and science level.
--:-- Host: I've reflected on whether science fiction becomes reality more quickly due to the exponential nature of technology. A book from the 90s might take a couple decades to enter reality, but because technologies are exponential and more people engage with them exponentially, that timeline might decrease. Do you have any thoughts on that?
--:-- Neil Stephenson: I think it's not just that it's decreased, but that it's swapped around. Science fiction is now trailing behind. Things are happening faster than science fiction can anticipate them, at least in the digital realm. When you're talking about big engineering—rockets, spaceships—science fiction is still way ahead of reality because those ideas take longer to implement. But trying to stay ahead of developments in the internet or online is a hopeless task. Many science fiction writers have talked about the fact that no matter how much they try to stay on top of trends and think ahead, by the time they write a book and get it published, everything they've written has already happened.
--:-- Host: I wonder when the first AI-written book will top the New York Times bestseller list. I wouldn't be surprised if that happens within my lifetime. But returning to carbon removal—let's close on this topic since it interests you. Given that you've dived so deeply into this arena, what are some of the most exciting, optimistic things you've learned? Or certain companies attacking this problem that you think are exciting?
--:-- Neil Stephenson: I'm excited by Terraform Industries, a company in Southern California founded by a friend of mine named Casey Handmer. He's a physics PhD who worked at JPL and that kind of thing, but decided to address this issue. They're developing a system that uses photovoltaics to extract carbon from the atmosphere and turn it into synthetic fuels. It's a net-zero carbon exchange because you're pulling carbon out of the atmosphere, selling fuel that gets burned in an airplane engine, and putting the carbon back. But it's still a big improvement on sucking it out of the ground and adding new carbon.
--:-- We've all been conditioned to think that anytime you use energy, it's bad and you should feel bad and curtail your use. That's only true if it's energy produced by burning fossil fuels. Energy from renewable sources, particularly photovoltaics, isn't hurting anyone. The cost of making photovoltaics is coming down—it's come down a lot and is still going down. I think we'll end up in a situation where we use energy from those sources to get carbon out of the atmosphere and keep it out. That's an optimistic trend I hope continues.
--:-- I hope it doesn't come to a billionaire having to operate unilaterally and release sulfur into the atmosphere, which is the premise of *Termination Shock*. I suspect that will happen, but it'll be done by some country, not a lone billionaire. A lone billionaire makes for a better novel, but a country is more realistic. When you look at the threat of so-called wet bulb disasters—where temperature and humidity become so high that humans can't survive—we could see very large fatality numbers in certain parts of the world. Why wouldn't such a country want to put sulfur into the stratosphere to cool things down? It's a tourniquet. It's the emergency thing you do on your way to the emergency room to get a real cure. The real cure is going to be carbon removal.
--:-- Host: It doesn't solve the problem. It's a band-aid until you remove the carbon from the atmosphere. Neil, I want to say thank you for taking so much time throughout this interview. You've inspired many people with your work, and it's important to note that you are a builder in this space. You have been for many years—at Blue Origin, Magic Leap, and now Lamina1. Many people are excited to see what Lamina1 brings, whether it can build a better ecosystem for creators, and whether we can use these design decisions to make a more interoperable, fair, and equitable ecosystem. Thank you again for your time.
--:-- Neil Stephenson: It's been my pleasure. I really enjoyed the interview. Thanks for being so well prepared and asking good questions.
--:-- Host: I can only hope that maybe one day we get to play with your GOAT AR application. Is that ever appearing in the world?
--:-- Neil Stephenson: It ended up getting turned into demo code for Magic Leap. Its real value for the company was to provide a sample app and some documentation that other developers could use to make their own applications. But I don't see why you couldn't have them running around in a virtual space.
--:-- Host: Hopefully we'll get to experience that one day. Thanks again, Neil.
