Building the Infrastructure of the Metaverse

Announcer:

Today, on Building the Open Metaverse.

Jon Radoff:

The idea that our identity online and how we relate to other people as digital beings is becoming even more important to a huge swath of the world than even our physical identity.

Announcer:

Welcome to Building the Open Metaverse, where technology experts discuss how the community is building the open metaverse together, hosted by Patrick Cozzi from Cesium and Marc Petit from Epic Games. 

Marc Petit:

Hello, I'm Marc Petit from Epic Games talking to you from Los Angeles today, and my co-host is Patrick Cozzi from Cesium.

Hey, Patrick.

Patrick Cozzi:

Hey, Marc. How are you?

Marc Petit:

Good, how are you? I'm enjoying California this week.

Patrick Cozzi:

Very nice. Yeah, I'm doing well as well. We're here recording on Monday, and I actually ran a half marathon Saturday morning, so I'm fully recovered. Feeling 100%.

Marc Petit:

Wow, that's crazy. Today we're very excited, because we have another pioneer on our show. Somebody who's a longtime serial entrepreneur and author and advocates for game developers. It's Jon Radoff, the CEO of Beamable.

Jon, welcome to the show.

Jon Radoff:

Thank you for having me, Marc. Excited.

Marc Petit:

We've been waiting for this moment because you've been talking about the metaverse way before Patrick and I. We're a little bit anxious today. Your blog has been something that's been very insightful for a long time. Thanks for that contribution; we're trying to follow your lead and be as rational and as good as you are.

Jon Radoff:

I think you just said I'm a hipster. I was talking about old-school, but okay, I'll own it.

Patrick Cozzi:

So, Jon, we love to kick off the podcast by asking our guests about their journey to the metaverse, and for you, you clearly have a huge passion for games and for programming. Let's go back to 1992 when you created one of the earliest commercial text-based massively multiplayer online RPGs, Legends of Future Past.

Jon Radoff:

I'm going to even go back further. The first game that I ever made, I was eight years old, and my father got me access to a mainframe computer at Digital, where he worked. I was also a huge nerd around Dungeons and Dragons at the time. So I made kind of a Dungeons and Dragons game that had 2D maps and fighting and things like that on it. I did part of it. I think my dad did a lot of it, but that was sort of my start in the industry. For better or worse, there's been sort of a role-playing game and D&D aspect to my career, there's been a computer programming aspect to my career, and I've run with it ever since.

By the time I was 19, I had been playing these MMOs; they weren't even called MMOs at that time. They were MUDs on commercial services like CompuServe and Genie. I met my future wife in a game called Gemstone, and we were just convinced we could build a better game than the one we were playing. So we ended up moving in together. I dropped out of college, and we launched Legends of Future Past.

But in my view, that kind of game experience, the whole multiplayer aspect where there's a heavy social element like that, that to me is the start of the metaverse. Even Dungeons and Dragons without a computer before that was the metaverse, and everything since then is using technology to provide more immediacy, breakdowns, spatial barriers, breakdown temporal barriers, and get us in the room together in the imaginary world together with each other. I've been doing that my whole life.

Patrick Cozzi:

Very cool. I like the origin story there and the idea of how far along the ideas in the metaverse, how far back they go. Let's talk about some of the stuff you've done outside of gaming. So you worked on Eprise and then GamerDNA before you started Disruptor Beam.

Jon Radoff:

Yeah, I don't usually get a chance to talk about all my non-game stuff, but when the web was coming along it was very technical, hard to make websites. We take it for granted now, because you can just go to Squarespace or something and launch a website. But, in the early days of the web, it was hard to do, and the opportunity I saw was to make it really easy for people not have to know coding or servers or anything like that. So we built a piece of software called Eprise, and what it did is it did all of that for you. It made it super easy, and the funny thing you bring up is it's like the same pattern I see over and over again in any kind of creator economy.

You take the web, for example; the earliest stuff is hackers and programmers. They just build stuff, because they're willing to take the time and figure it out, and they make stuff, and it's way more work than it should be, but they do it anyway because it's fun. That's what we hackers do.

Eventually, what you need is something that anybody can access really to scale up across the market. Whether that was websites back in the day of Eprise, or whether e-commerce that's been democratized by companies like Shopify now or the whole era of 3D engines.

There are a couple of companies that have built really strong 3D engines that have opened up access to the whole universe of spatial computing and graphics to people. To me, the metaverse game development, generally speaking, is lacking that framework that makes it easy, so that you can imagine something, sit down in front of your computer and just go to work on the creative side of things without having to be so concerned about all the plumbing, the technology, scalability, economic system, all that other stuff that really makes this kind of software work.

Marc Petit:

And you took Eprise public, right? How was that IPO experience?

Jon Radoff:

Insane, crazy, interesting. Learned a ton. Some of the stuff I learned definitely remains pertinent to this day. Of course, in other cases, things have changed from the dot com era, but I started a company, and two years later, it was 20 million in ARR, which was amazing. We were able to really create something from it and build a public company.

It was a weird market when we went out because then everything went sideways for quite a long time, and we ended up merging with someone else. But great experience, got to build so much, work with great customers, solve real problems and see what happens when you go from zero to 300 people in two or three years.

Marc Petit:

Quite an amazing experience for an entrepreneur to go that fast to an IPO and then manage it through a downturn. Downturns happen, as we can attest.

Jon Radoff:

I've noticed.

Jon Radoff:

I've been through a couple in my career; it's always challenging.

Marc Petit:

Yeah. So tell us about Beamable. What were the founding principles behind that company?

Jon Radoff:

Well, between some of the stuff we were just talking about and Beamable, I had run a game studio called Disruptor Beam, and we built some games built on very popular TV shows. The biggest game we built was a Star Trek game called Star Trek Timelines. We also were the first online game for Game of Thrones. We made a game called Game of Thrones Ascent. So, we really cut our teeth on the whole experience of how you bring story and multiplayer and metaverse-y kind of social interaction together, but around these grand stories and universes that people love so much.

It was through that experience that I saw a few problems. One of them was that you spend so much time building the infrastructure and the technology and the scalability, and things like purchasing systems and social systems, all of the foundational pieces, which everybody always underestimates.

And even if they figure out how to build a few pieces of it, they always underestimate scale. Companies always run into problems when suddenly they have millions of users for the first time. I've seen companies get shut down because they had millions of users and they weren't prepared for it, which is a real tragedy when that happens. The whole idea behind Beamable, which was born out of a whole reorganization outside of Disruptor Beam, was to focus on the technology to really bring games to life.

We use this term live services. Live services is literally about bringing games to life for communities of players who are going to be interacting with each other, competing with each other in real-time, cooperating with each other, all of the social systems that go around that, living dynamic economies within games, and all of the customization that you need to bring to game servers and game systems to enable that. That's what Beamable is, and that's what we've been doing now with a number of games that have launched over the last couple of years with us.

Marc Petit:

Let's talk a little bit more about those online services. They're going to be a critical component of the metaverse because, kind of by definition, everything is going to be social in the metaverse. Where do you see the biggest need for innovation there? We still have a hundred players per instance. This is not very social. How do we break those barriers? How do you think companies like Beamable can help in that respect?

Jon Radoff:

Well, there's foundational stuff that is just still really complicated to incorporate in your game, and it's sort of basic table stakes like, how do you actually have a persistent world? How do you preserve the state of all your users? How do you preserve the state of all of the objects and items and things going on in your universe? Even that alone is wildly inconsistent from company to company, and usually, what they do is they buy web server technology. They'll use node JS or something like that, because that worked for websites, and suddenly they're building a game server off of web technology. We see that all the time.

The foundational piece of data store, the persistent world, and then the objects that you create on top of that, like users and their identities, their account history, and the competitive leaderboard as people start competing with each other or cooperating with each other. The social systems like Guilds, and cooperative systems within Guilds, how do you recruit people to Guilds, all of the economic systems?

How do you maintain all of the SKUs, so to speak, to use e-commerce language, but all of the items, the things that work within your environment, how do you relate that back to how you obtain those items?

How do you purchase them? Do they spring forth from treasure chests that you find along the way?

These are all the things that people end up spending like 70%, 80% of their time building if they end up trying to build all that stuff themself in a live game, instead of the actual thing that's important for a game developer, which is storytelling. I spent a lot of my career with story-based games and Star Trek and Game of Thrones; all we really wanted to do in those games was focus on, what is the core fantasy that you have in those worlds? How do you deliver that to the player?

We did a pretty good job in those two games, but we didn't do nearly as much as we would've liked if we weren't building in-app purchase systems and account systems, and data store systems that needed to scale to millions of users.

Patrick Cozzi:

So, Jon, I appreciate that you present yourself as someone who fights for game makers and was hoping you could share with us a bit about how Beamable helps play a role in that.

Jon Radoff:

Thank you for invoking the mantra of our company. That's the culture of Beamable, by the way. That's something I put in my tagline, but it's something we tell ourselves every day. It's on all our materials. We fight for the game maker.

Game-making is so damn hard. Full stop, that’s the basic problem. Game-making is so hard. There are so many elements, there are so many things that can go wrong. Building the right team for a game is challenging. Figuring out how to capture the fun, but then not only capture the fun and build a fully comprehensive system around it, then figuring out how do you engage a customer over the long term. And then, lastly, how do you connect with an audience? How do you even find the audience, like user acquisition, and find a way to scale that? All of that is so, so hard.

When I talk to game developers, and I've been a game developer and still feel like I'm a game developer, even though it's at the tech layer today; what everybody really loves doing, number one, is making a great game. It's going, again, to the storytelling, the experiential aspects, the graphics, the artistry, the feature set, and the engagement loop of the game. That's what we all like actually making, but we don't get enough time on that. So when I say we fight for the game maker, it's really to fight for that person who cares about that list of things that I was just describing and make sure that they can spend as much of their day as possible on those things.

Because not only is it fun for us as game developers, that's why a player is going to buy a game. Right? If you think of it as investment ROI terminology, all the alpha that you're going to deliver in the ROI of your game is going to come from how fun the game is. All of the risks you could potentially contribute are going to come from things like technology, scalability of the tech, and scalability of user acquisition. Those are things that I think just need a lot more specialization and stable platforms that people know they can trust, rely upon, and focus on the craft of game-making.

Marc Petit:

Recently, the news about the merging of Unity with ironSource; they’re kind of juggernauts in this space. How does Beamable position itself in that landscape versus those bigger guys?

Jon Radoff:

We think conceptually of the universe of technology that you need to deliver a game fundamentally comes down to the 3D engine to deliver the experience. Then there's an enormous amount of live services infrastructure, right? The live services infrastructure is super fragmented today. There's no consistency.

Earlier, you were asking a little bit about what differentiates us. What do we see as the big problems in the marketplace? Well, a big part of it is just having a workflow system that a game developer can sit down in front of and select from the kinds of live service elements that they can build their business around. Not unlike the way you can sit down now in a 3D engine and actually build worlds, build graphics, synthesize all the different pieces, what I call composability.

So the composability of 3D graphics and world-building and the experiential aspects of games today is really outstanding compared to what it was a decade ago. In fact, it's improved tons just in the last couple of years. But that same composability, the ease of workflow, the ability to just drop something in and expect that it will scale, that hasn't been accomplished amongst all the live services elements. So that's our focus, is providing that framework around it, the workflow that makes it really easy to incorporate live services into a game.

Marc Petit:

I think we have a lot of shared customers between that, and the knowledge that you bring of games, I think, is pretty unique. This is a deep understanding of game makers, I think, which is the hallmark of Beamable, if you allow me to say this. I know you would not say it yourself.

Jon Radoff:

That's kind of you.

Marc Petit:

We also hear a lot about new technologies, such as blockchains and NFTs, in the world of gaming. Of course, these are more concepts that would be implemented at the back end. Right now, it's the crypto winter. I mean, it's not as fun to talk about NFT games, but do you foresee a natural usage, now that the dust is settling, or we're starting to understand better this technological landscape? What's working, and what's not working? Is that something that inspires you there?

Jon Radoff:

Well, if we take a step back, let's talk about what I think remains interesting about it, because certainly, it's a complete mess right now, and the market, unfortunately, is full of some really bad actors and people that are just in it for the whole financial speculation game without really being concerned about the ultimate value proposition that it contains. I think you can identify a few things that are interesting.

Number one is, really, blockchain is a way to solve the problem of consensus between lots of computers. Consensus is a hard problem to solve unless you're all willing to trust one central authority who just stores your data and then tells everybody what they need to know. I mean, that's how the world works today, everything is a trusted authority, and that's okay if you're willing to live in someone's ecosystem, number one.

The challenge with it is getting all the other big players with their own ideas about how they would like to own those ecosystems actually cooperating with each other. Blockchain, in terms of solving a consensus problem, does it pretty effectively in terms of being able to have big parties as well as any smaller developer who wants to participate in a common dataset without having to say, "Hey, this one particular company is going to be in charge of everything and have all the keys to the kingdom."

Now, over the last year, in particular, one of the previous critiques of it, was, "Well, okay, but consensus is really hard to do with proof of work algorithms because of how much cryptographic protocols consume in computing power." I mean, that still remains true for certain things like Bitcoin, but for the things that games and metaverse-type stuff will actually run upon, it's moved on to these staked algorithms, which don’t require nearly as much energy. It's a 99.9% reduction. But I think solving the consensus problem between lots of parties is really interesting.

Then within consensus, the thing that then really opens up, I think, is the idea of a programmatic exchange of value between parties. Again, without requiring a broker in between or someone who owns the whole data store in between. The ability to have a piece of software that says here's some value, meaning money, assets, et cetera, that can exchange with another piece of software is extremely powerful in terms of composability.

We were talking about composability earlier in terms of the universe of 3D graphics and how much more composable that is, the lacking of composability in, say, live services. Well, part of that is you have game economies, and you would ideally, for certain kinds of things, especially some of the metaverse-type stuff that we've been talking about over the last year, need composability of the economic systems within these universes. That's where I think blockchain gets interesting. But, yes. I think the market is currently retrenching, and there's been tons of unfortunate behavior in this and lots of scammers, and it's been disruptive, to say the least.

Marc Petit:

Yeah, I agree. We're starting to see some use cases that are really valid, and it will be interesting to see how those technologies really get adopted and implemented.

Let's switch gears and talk a little bit about the metaverse as well. We don't quite know what that is, but we like to talk about it, and you've been writing extensively about it, too. By the way, if you don't follow Jon on Medium, you should. There's a lot of interesting content, and I reread some of it because it's funny to read something from 18 months or two years ago, and your stuff is holding the test of time pretty well. Things change so rapidly.

You introduced this concept of layers of the metaverse. Can you talk to us about layers or pockets of the landscape where you're seeing the most innovation or the stuff that excites you the most in the metaverse?

Jon Radoff:

Yeah. Let me even just take a step back from a moment, because you did raise the whole subject of what does metaverse even mean? There are different ways people talk about it. To some people, it is the crypto stuff we were just talking about. I think that can be part of it, but it's not equivalent to metaverse.

For some people, it's AR/VR, embodied experience. That's kind of the Facebook version of things. To other people, it's virtual world platforms. It's Roblox, it's Fortnite, it's things like that. So I think that there are aspects of truth to all of those, but the way I've tried to think about these technologies through my whole career, going all the way back to Legends of Future Past, which we led with earlier, is there's a culture shift. There's a social shift underway, and I think it's really critical to understand how people are using technology differently today and how that's been shifting over time.

The shift that's happening is the idea that our identity online and how we relate to other people as digital beings is becoming even more important to a huge swath of the world than even our physical identity. So I think I was maybe leading the way a bit when I met my future wife in an online game. When I did it, by the way, that was very weird to do. Probably for a lot of people listening to this now, far less weird. You've been scratching your head wondering what I'm talking about, why I think it's weird, but it was weird when I was doing it.

I think if you look at that trend over time, what you're seeing taking place is exactly what I was describing, which is that people are investing more and more in their digital identity. And when you start with identity, you then extend out to that into your creativity as well, so your creativity that you express, maybe first through your avatar, through your socializations and your social groups, like guilds and online games and eSports and performance, all that kind of stuff is the next step. Then, ultimately, it will be people shaping and crafting worlds, not unlike what they do in Minecraft. I think of Minecraft, for example, as legitimately part of the metaverse. I don't have these strict barriers between metaverses. Are we there yet? I think we are there because of the social and cultural trends.

To answer your question about the layers, I have this whole seven-layer model where I've tried to break down the value chain of the industry, and what feeds into the next, but I'll just kind of focus on the opposite extremes of the seven layers, and then comment briefly on what happens in between.

Ultimately, the only thing that most people are going to care about with respect to, quote-unquote, metaverse, or whatever we end up calling it in the grand scheme of things, is the experience you have, right?

The experiences are almost entirely games today, but that same game experience, the craft of game-making that many of us have learned to do, is being increasingly applied to things outside of that.

It will impact things like shopping. It'll impact things like simulation in virtual spaces. It'll impact things like the experience of music. Just like you can go to a concert in Fortnite and Roblox and join tens of millions of other people who experience a concert, well, that dialogue that you have between that performer, in that case, music, but it could be any kind of performer, and you in the audience is being writ large on the metaverse. That's really what the experience is about.

On the opposite extreme, there's incredible innovation happening at the very foundational levels of technology. So the speed of networks, the speed of semiconductors, the whole revolution around GPUs, which I think of as from a technology perspective, this is really the matrix generation. I'm not talking about Matrix, the movie, which, although that's funny to think about, matrix in terms of just matrix operations.

The GPU does two things amazingly well. It does more than two, but at least two things really well. One is the spatial computing applications of matrix math, and the other is the ability to put artificial intelligence algorithms and train models, and run AI models through matrix technology. GPU does that because, of course, it can do matrix operations in parallel at huge scale that we couldn't do on CPUs before.

We got experience on one level, which is just delivering those experiences. The GPU kind of tells you where we're going because it's giving us richer, more immersive, spatially computing-oriented environments. Whether that's on a screen, whether that's in a VR headset, whether it's in some future AR goggles, it's sort of all enabled through spatial computing, but also AI playing a bigger and bigger role along the way as well, which is everything from characters that you're going to interact with within games.

Games are the ones that have probably done more in terms of characters with NPCs now for many decades, and these characters are going to get more and more interesting. It's AI being applied to the whole creative process itself. In the last year, I think people have been blown away by all these computational creativity products that have come along. I'm talking about Stable Diffusion and stuff like that. The ability to just take a text prompt and turn that into usable material. I think we're going to see more and more of that stuff helping with the creative process.

So there's a huge number of AI applications there, but I'll end the layers without spending an exhaustive amount of time in the five we didn't talk about a lot. The middle is really about unlocking creativity, whether that's the creator economy itself, whether it's about the tooling, whether it's about mass market acceptance of the hardware that you need, whether it's the use of blockchain, for example, or open source as a consensus layer or decentralization layer. All of that is about unlocking creativity so that you can deliver the experience to people.

Patrick Cozzi:

So, Jon, one of our favorite topics is interoperability and open standards. It comes up on every episode of the podcast, and Marc and I have organized a few SIGGRAPH events, and it comes up over and over again. This is a topic that you've written a lot about as well. You have a great article on the layers of interoperability, and you defined five layers: connectivity, persistence, presentation, meaning, and behavior. I was hoping we could talk about these a bit, maybe starting with connectivity, which I believe you believe is a mostly solved challenge at this point.

Jon Radoff:

Well, it has to get a lot faster, and there are a lot of super interesting problems. I don't want anyone in the 6G world to reach out to me later and be like, well, we're still working on that. Yeah, no, there's really, really, really hard problems.

Patrick Cozzi:

Understood.

Jon Radoff:

Let's take a step back again on interoperability. This is where people get caught up on interoperability. I think people sometimes get trapped into thinking that interoperability has to mean this vastly monolithic system in which everything you can possibly do is prescribed for you, and you have to operate within a very distinct set of constraints. I don't think interoperability means that, right?

If you go across those lists of interoperability domains that we were just talking about, well, there are ones where that makes a lot of sense. For example, the connectivity layer. I think you could reasonably argue that TCP/IP is an amazing interoperability layer that beat out a lot of proprietary networking protocols that existed in the past. Today you can plug your computer in and gain access to all kinds of services through TCP/IP.

Protocols are kind of the foundational layer of the metaverse. That's why another way I think of the metaverse is it's really just the next generation of the internet building upon these things that already exist, but adding more aspects of creativity and spatial computing and real-time connection with each other.

Patrick Cozzi:

Let's talk about the presentation layer, interoperability there.

Jon Radoff:

As you go up this chart, which you're referring to, it goes from stuff where it's a little easier to define very specific ways to plug in interfaces between, say, hardware and software layers and TCP/IP at the connectivity layer, but it starts getting squishier as you go up. We have the world wide web, for example. The web is a way of standardizing a huge amount of how we deliver the presentation layer. HTML is a presentation layer.

HTML isn't amazing for delivering things like 3D immersive experiences. So people have come up with systems, actually really impressive systems, that use things like JavaScript to do that. There's really interesting work happening in things like web assembly. There are all kinds of solutions that have been built on lower levels of languages and the basic technology of a web browser to deliver the presentation layer.

I'm personally, a big fan of the idea that accessing the metaverse in the future is going to tap into a lot of things that come from web technology. Now, the web has to get more real-time. There are a lot of problems we have to solve. A lot of the code has to be much more easily embeddable, maybe stuff like web assembly that I was referring to earlier is one of the solution pathways for that, amongst others that people are looking at.

I think that it's got to be like a browser, whether it's a web browser or a meta browser or some other things that we define in the future. It's got to be a browser-based technology that allows you to connect to any kind of service and access it, whether it's an MMORPG, whether it's a shopping experience, whether it's that music concert that we were just referring to earlier because that's what will really dramatically expand access to all of this stuff.

What goes hand in hand with the presentation layers, if you can start defining that, how you actually render and deliver it, you want to make that readily accessible to the authoring environments as well.

There are a lot of people working on how you standardize that as well. Just like we had authoring tools for the world wide web, and then it eventually migrated towards actual online tools like Squarespace or Shopify and whatnot, where you could do it inside your web browser.

More and more of that creative process at the visualization layer, the presentation layer needs to just become a lot easier through the tooling of it. Could be universal, like USD, for example. We don't know what the standard is, but we need more of those frameworks defined so that more tools can enable the creators to deliver through a common browser-based interface what the actual experiential layer is.

Marc Petit:

Yeah, I think we're in agreement with that.

Patrick and I are part of an effort, the Metaverse Standards Forum that you recently joined, and it's really about understanding that presentation layer and trying to understand if what we've seen from USD is very prominent in authoring tools and, as demonstrated by NVIDIA, very prominent in the runtime space. Could those concepts be the foundation for that new presentation standards, which is akin to HTML, but working for the 3D virtual world?

It's kind of something that's very much top of mind for many of us right now at the Metaverse Standards Forum, and trying to validate that hypothesis.

You joined the Metaverse Standards Forum. What are your expectations there?

Jon Radoff:

I'm approaching it with a lot of humility because there are so many people there that are smart, and I think everyone has different views of how you deliver these applications.

That's why it's important for me to try to break down these other areas of interoperability, because that's where I find people tend to grind into analysis paralysis because interoperability has to mean everything from, I don't know, USD to define the way you describe the graphics of a world and the object placement in the world.

If that then also has to capture everything that those objects could ever be, including from a behavioral standpoint, and from an economic standpoint, it just gets really super hard. I'm interested in really focusing on how we draw boxes around that and identify ways where cooperating groups of people who are building gains or metaverse applications can find those areas of agreement and work within larger and larger frameworks.

Conceptually speaking, TCP/IP just allows everybody to cooperate and not worry so much about the network layer anymore. You can just build within it. Very few people these days building an online game worry too much about it. They just use what's already off the shelf. Then the fact that you've got a couple of really great 3D engines out there and you've got, potentially, the ability to define a common and consistent way to deliver the presentation of world space, that helps out a lot.

But when I talk about things like behavior in the world, it's not just physics, for example. It could be like, what are the game rules that are associated with that object? How do you define those game roles? How do you make it easy to interpret process and execute and force the rules about a particular game object, for example, in a way that's super scalable and just works?

These are interesting problems to solve, but you don't have to solve everything at once. You can break it down and create interfaces between them, and not everyone has to agree, by the way.

This comes back to the blockchain. When I've talked about interoperability, the use of blockchain to provide an economic backplane for the exchange of items between different experiences, the most common feedback is, well, people have tried those kinds of things, and they've never really worked. Those people haven't really played something like Roblox, apparently, where people make all kinds of games and exchange items, and they go between lots of different experiences.

There are ways to achieve constellations of economic interoperability between games.

If I'm within a cooperating group of game makers, we could all agree that we'll allow a certain kind of item to go between our worlds. Just like you have that literally right now in Roblox.

That doesn't mean enforced interoperability. It doesn't mean that that item that exists in one world has to go to other worlds. You can always man the gates. You can run your own theme park and say, here's what I'm going to allow in; here’s what I'm going to allow out. This is, to me, real interoperability, which is allowing people to agree on the ways that they'll work with each other and have composability and allow smaller teams to do really interesting work where if they had to build an entire platform for virtual worlds, they would never get to the actual cool idea about the experiences that they want to build.

Marc Petit:

I think leaving the people who create the world and setting the rules is important. I don't think to mandate and force interoperability of everything into anything.

Jon Radoff:

Interoperability doesn't mean that I go to World of Warcraft and my costume makes it into League of Legends. No one's trying to force game developers to do that. I think people want to make the option for individual game developers if they want to do that and decide how you render it, what that means, what it means socially, and what it means economically, and allow groups of developers to work together the way a few Roblox developers do that today.

Marc Petit:

This is actually a very interesting segue into interoperability, the means to support a business model. We all aspire to this creative economy, and you've been writing extensively about that as well. We had Philip Rosedale on the show a few weeks back, talking about the second life business model, which was very simple. You pay your fee for being there, some sort of property tax, and then you had a little bit of the VAT tax for the exchange of goods within the world, and that was pretty much it. Something that's a very, very simple bottom-up economy, it seems to work very well for Second Life. We're not seeing anything similar in the Roblox, or all the Fortnites of the world. Do you have a view on what would be a natural business model or economic model for the metaverse?

Jon Radoff:

The first comment I want to make about business models, in general, is this is why we need to allow for experimentation on all kinds of business models.

What I'm actually getting is there are a lot of platforms today that have big taxes associated with them. We know what platforms I'm talking about. They take a huge portion of revenue, and they pretty much require that you have certain kinds of business models to be viable, either an in-app purchase model or an advertising model. There's nothing wrong with those models inherently, but it's very constrained, and we're not seeing innovation on all the other things you might do.

This is why I still love PC games. As much as I use my mobile device all day long and I've got all the consoles, there's pretty much every device you can imagine from a gaming standpoint in my house, but I love PC because it's the platform where you can still pretty much choose to do whatever you want as a game developer. You can create your own business models.

The web is like that too. The web has struggled to deliver the game experiences that people want, but the web, you could almost think of it as an extension of the whole PC gaming ethos because it's open and unconstrained.

I want to see more of that. It's really important not to tax innovation, and then expect innovation to happen. That's sort of my high-level thesis on it.

In terms of a business model specifically for metaverse, one of the things I look at, rather than tell you, "Here's the formula for charging people to use your metaverse," I know people are experimenting with things like land, not too dissimilar from what had been tried in Second Life and trying to apply that to all these other newer metaverses that people are currently creating. I think it's interesting to look at, what are the jobs that are actually going to be formed in the metaverse? I think it's pretty interesting to think about real-time activity, and real-time interaction. You look at music performance.

Now, it's been done with tens of millions of people and things like Roblox and Fortnite, for example. But I'm really interested in seeing how that scales out across a whole market where every concert doesn't need to be for tens of millions. Maybe it's for a small group of people. How do you deliver that experience? How do you bring live performance to life across more and more of these applications? We already know that there's evidence for that.

Not just the music I was referring to, but if you just look at eSports, if you look at streamers, there are so many things that could be the fusion between AI, live performers, people doing things, new avatar systems, spatial computing, new forms of creativity where I actually deliver an experience to you, practically in real-time. Not unlike LARPing and dinner theater experiences, all of that stuff. We could tap into a whole new class of jobs that are performance artists in the online world.

I look at that. I think of it as really opening up creativity, though. Some of it will be performance, and some of it will be more bespoke, like the crafting of avatar costumes, and the crafting of worlds. The more and more and more we can really open up the creative space, so that if you can go direct from imagination to the screen, or whatever it is, that we can cut down that loop as much as possible, then you'll start to see the emergence of more of those jobs. And as people do the jobs, that is kind of how you're going to prove many of the use cases, and you'll start to create things.

Before we had Twitch, we had Justin.tv. It was basically performance art, and then it became Twitch because we were proving that the job could work. We need the ability to allow people to do those jobs and not get in the way of the business models that they could employ. We'll discover what the next generation of business models in these worlds will be.

Marc Petit:

We all aspire to a more transactional business model, I think, but we recently heard that Roblox is delving into advertising. Do you think it's a sign that the economic model needs advertising to be sustainable?

Jon Radoff:

Any kind of experiential product is really monetizing attention, at the end of the day. Advertising is a good way to monetize attention for a certain class of content.

For content where there are not a lot of tremendous incentives, say to make an in-app purchase and buy something, or where the incentive to do so is very, very low, something like advertising can end up becoming a more efficient way to monetize that attention. Whether it was putting quarters in the arcade machine years ago, buying the sequels to a franchise, or buying DLC as it keeps coming out, everything is really back to attention. Advertising's totally legitimate. There's a whole body of games and content where advertising is just going to be the best way to do it.

What I would hope, related to the earlier statement I made though, is I hope we don't just converge on one business model that everybody has to do.

Mobile games, for example, you kind of have to build IP transaction-based games today because it's just not really economically viable unless you're in pretty specific use cases for advertising or part of big, big content networks where you're constantly sending the user from one hyper casual game to the next.

Largely, it's still an IP-driven business model, but that's the way the system, intentionally or not, was designed. It comes back to giving people the flexibility to experiment and try things, charge for things directly, and come up with new subscription models on their own. There are so many different ways to approach business models, all of which will relate, in some way or another, to attention capture without telling them what they have to do.

Marc Petit:

For me, advertising is responsible for a lot of the issues we're seeing with the current mobile platforms. That's why I always wonder if there is a way we can do advertising and it doesn't take the prominent space, and drive the behaviors of everything like we've seen in the past 15 years. So that's an open question in my mind.

Jon Radoff:

I think that things are going to get more community-oriented, more social over time. It's amazing to me, for example, that Discord doesn't try to just sell me games directly that my friends are playing. I mean, I'm just riffing off the idea, the fact that I can see what my friends are playing and immediately get some insight into what's fun about it and make a purchasing decision. Stuff like that, I think, is going to become more and more common.

If it's not quite as a direct, I described a direct response model, but even in cases where it isn't direct response, the fact that people have social organizations that they play in, like guilds and clans, and whatnot, making the social groups more transportable from experience to experience is something that some games would benefit from, on the incoming side at least.

I think it's interesting to think about a metal layer that wraps around a lot of games that allows that kind of transportation.

We already have people, again, back to performance like eSports and streamers, and whatnot; they have their own communities. So thinking about how you take those communities and you intersect them with the whole way games are propagated, not necessarily through advertising and sponsorship, which clearly they already do, but I see companies starting to think about more creative ways of interfacing shopping experiences and introductions and things like that.

Marc Petit:

Yeah, makes a lot of sense. I'm optimistic, too.

Patrick Cozzi:

Jon, we covered a lot of good stuff today. First, thanks for sharing your passion for developers and game developers, but it was great to talk about infrastructure to enable developers and creators, defining the metaverse, the breadth of interoperability topic, and everything that you and Marc were just talking about around economics and business models. As you know, we like to wrap up the episode with a shout-out. If there's any person or organization you want to give a shout-out to.

Jon Radoff:

Well, certainly, I got to give a shout-out to my team back at Beamable. They're amazing, and they can really help you build a game that's a live game.

I'm going to go back to where we started the whole discussion today. I got to give a shout-out to my parents because it was my father who plopped me in front of that VT100 terminal when I was eight years old and set me on this path. It's been an incredible career to be able to make games and work with game creators and build online technology. So thanks, dad.

Marc Petit:

Thank you, Jon. That was fascinating. You are a true gamer and game dev at heart; that's why we love this community so much. It was fantastic to have you. Jon Radoff, the CEO of Beamable, also a blog author on Medium called Building the Metaverse. I strongly encourage people to follow you, and read everything you've been writing about games at the metaverse. It's been fantastic to have you.

Thanks. Thank you very much, Jon, for being with us today.

Jon Radoff:

Thank you so much.

Marc Petit:

Patrick, thank you too. Thank you to everybody who is listening. We always like to hear your feedback, like to hear your suggestions, and your critiques. Please hit us on social, let us know what you think about the podcast. Thank you, everybody, for listening. We'll be back with a new episode in a few weeks. Thank you very much.

Patrick Cozzi:

Thanks, everybody.