The State of the Metaverse

Announcer:

Today on Building the Open Metaverse.

Matthew Ball:

If you believe that the metaverse is a multi-trillion dollar opportunity, if you believe that it is a next generation version of the internet, then you have to believe that billions of people will participate in the metaverse on a daily basis. If so, the point in time in which we're sitting right now, we're existing, is kind of like the ICQ era of online, or perhaps 2005. In 2008 even, Facebook had 150 million MAU's, monthly users. In 2009, they had 350 million monthly active users. A decade and a half later, not even, they had 3 billion MAU's and 2 billion DAU's.

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 everyone, and welcome to our show Building the Open Metaverse, the podcast where technologists share their insight on how the community is building the metaverse together. Hello, I'm Marc Petit from Epic Games, and my co-host is Patrick Cozzi from Cesium. Hi, Patrick. How are you?

Patrick Cozzi:

Hi Marc. Hi, everybody. Doing great.

Marc Petit:

So I missed the last episode and you went full geeky on us with the WebGPU.

Patrick Cozzi:

Indeed I did.

Marc Petit:

So I have to learn to not let you alone, and so today we're kicking off season two of our podcast series, and first of all, I'd like to take a second to thank you all for listening. Our first 19 episodes have been listened to for total of 32,000 times, and we have more than 2,400 subscribers and engaged listeners. Not quite sure what those numbers mean, but they're for sure motivating for us to continue. All right Patrick, so we're moving on. We're keeping, doing that thing, right?

Patrick Cozzi:

For sure. For sure.

Marc Petit:

And we have a few goals with this podcast and hopefully we're tackling some of them, like we would like to be an antidote to the metaverse hype and bring to light the hard work that has happened in the computer graphics industry for the past 30 years and that's now enabling the metaverse of today. We are here because of all of this, 30 years of research and development, and the other thing is both the visual effects community and the game developer community have a track record of collaboration in coming together to do what's right for the community, like we see in Khronos, like we see in open source, open standards, and Siggraph and even GDC and all of those events where we come together. So that's what we're trying to do here. As we are contributing to building the metaverse, we want to make sure that we highlight and celebrate those initiatives and carry that spirit into the creation of these new generations of platforms.

Marc Petit:

So that's what we are here for. Hopefully you guys need to tell us, give us your feedback, and we succeed in that celebration, and again, carrying the spirit into the future. So after a string of rather technical topics, we'd like to take a step back, see where we are with the metaverse, and put things a little bit in perspective before we jump back into technology and the hard topics of making and keeping things open and accessible. So we decided to go back to our roots and invite Matthew Ball again. Matthew was there with us for the first episode. So Matthew, welcome back to the show.

Matthew Ball:

Thanks for having me. It's been a delight listening to the podcast over the last several weeks. I'll say that I'm probably a few hundred of those listens personally, but I've certainly learned quite a bit and thank you for your efforts.

Marc Petit:

Well, thank you. So if you know about the metaverse, you know about Matthew Ball. But just in case, do you mind summarizing for us who you are and your journey to the metaverse?

Matthew Ball:

Sure. So Matthew Ball, I run a HoldCo/OpCo called Epyllion. We do a number of different things. There's a small, early stage venture fund, a venture advisory, corporate advisory arm. We produce the index behind the Roundhill Ball Metaverse ETF that's on the New York stock exchange, and then I have a few different entities that produce TV and film as well as games. But for the most part, organized around this theme right now, the metaverse.

Patrick Cozzi:

So Matthew, I think you're doing a fantastic job tracking the metaverse and educating the community. We'd love to hear what you think the current state of the metaverse is and if you think we're in a hype cycle, and if so, where in that hype cycle?

Matthew Ball:

So we're certainly in some form of hype. When you take a look at all of the core metrics. Before Roblox filed for their IPO in October of 2020, there were five ever mentions of the metaverse in an SEC filing. In 2021, there were over 240. Over the preceding 20 years, there were fewer than one dozen mentions in any Bloomberg terminal, an indexed article saying metaverse last year, there were over a thousand. Year-to-date in 2022, we're at over 150 mentions of the metaverse in an SEC filing. We've already eclipsed last year's total, and yet we see two very significant shifts that show that we have past that initial crest or peak. Number one is all Google trend data around the term metaverse has roughly halved over the past two months, and that correlates to the consumer response, which is after several months of being told the metaverse, the metaverse, the metaverse, a lot of consumers are saying, what are the products in the metaverse, or please tell me how our world is different than it was a year ago.

Matthew Ball:

And that's an important point for us to all pause and reflect on, and this really reflects three different things. Number one is differences of opinion as to when the metaverse is here, or perhaps if it's here already. Bill Gates and Satya Nadella have said that it's here. Jensen Huang just talks about it emerging over the coming decades. I believe that that's Tim Sweeney's perspective as well. Mark Zuckerberg and John Carmack at Oculus say that it's almost here, but it's five to 10 years away. And so this mixture of the suddenness through which the topic came out, the difficulty in understanding exactly what is different, as well as the disagreement as to whether VR is critical or peripheral is leading to this kind of doldrum where everyone's now trying to reassess was the hype too fast? Was it too sudden, or perhaps is this not even real to begin with?

Marc Petit:

And so we saw some financial impacts. So tell us about your tracking index for the metaverse that you created almost a year ago, right?

Matthew Ball:

Yeah, that's quite right. So my company produced the index, which we then licensed our company called Roundhill Investments, which to turns it into an ETF. The goal of the index was to bring together a council of experts, the former co-executive producer lead game designer for Red Dead Redemption, Grand Theft Auto V, the former head of developer relations at Oculus, former hardware lead at Valve, other individuals in this space to say, "Let's come out with a methodology to track the economic value or public equity side of the metaverse's evolution," and the goal there in abstract sense was to provide the everyday individual investor a simple way to gain access to this potentially multi-trillion dollar opportunity without needing to familiarize themselves with every single company and their relative exposure. At the same time, the stock market is what we call a popularity machine in the short term and a waiting machine over the long term, which is to say over time it proves itself based on its worth, but in the shorter periods of time, speculation and other acute events can distort things.

Matthew Ball:

And so the index, if we said that it was at about 100 at launch or normalize it, that was June 30th, we essentially got to 125 by mid-November. Mid-November was the peak of the metaverse frenzy as a narrative with investors. Facebook changed its name, Roblox and Unity, they surged to more than twice their IPO price even though they had only been public for about a year or six months, and since then that index has come down from about 100 or a peak of 125 down to about 70%. That's not atypical. Most tech stocks, most foreign stocks, anything in the so-called growth era, or areas, have kind of cooled, and that reflects exactly what I mentioned which is an understanding that if the first step is awareness and excitement, the next step is settlement. It's investment, and it's waiting for the products that will change the world.

Marc Petit:

So what's your foresight to the... You think it's a natural cycle of things?

Matthew Ball:

Yeah. It... Look, it is the natural cycle of things. I think there are two or three different ways in which we could break it down. Number one is to recognize that the gaming industry by and large goes in these five to six year bursts of really significant innovation. We saw that with MMO RPG's. Over the past five or six years, we've really been experiencing the ascendants of the Battle Royale genre, which was concurrent with the rise of these virtual world platforms, your Roblox and Minecraft. It's actually been a little bit of time since we've seen a massive discovery in what gaming or 3D rendered virtual socializing looks like. And so we're probably due for one of those big creations over the coming years, but right now we're going into a quiet period. Most metrics show that. We know that PUBG is down probably 30, 40% of its player base, Free Fire is down about 40%. Activision has talked about the fact that Call of Duty War Zone and Call of Duty mobile are significantly lighter.

Matthew Ball:

And so one element is we're waiting for the next big leap in gaming. The aiming isn't the metaverse, but it's clearly the forerunner. The second is a lot of attention is being placed on AR and VR devices. It's easy to say we've been here and VR didn't pan out. This was 2016, 2017, but it's easy to overlook how much better these devices are. Their frame rates have doubled, their resolutions have doubled. So we're looking at roughly four X the number of rendered pixels per second while we're also seeing new sensors put in the device, and we're seeing better graphical quality while improving battery life. That's an enormous leap, as you well know, and yet we remain pretty far out from actually having what most people consider to be MinSpec. So we generally believe that VR devices need a 50% hike in frame rate, yet they probably need a doubling in the resolution, and while I just said that we've done all of that in the past five years, these things are a little bit like GPA's.

Matthew Ball:

Going from a 3.4 to a 3.8 is much, much harder than... Or sorry, much much easier than going from a 3.8 to a 3.9, and so it looks like we're going to have to spend potentially a decade getting those devices farther still. The last big area as for what's predicted is where we talk about the ancillary or new applications of real time rendering engines. This is where we talk about virtual production in TV. I get most excited about the ongoing deployment into AEC, or AEC, Architecture, Engineering & Construction, and in the fall, John's Hopkins began performing their first ever live patient surgery using AR and VR devices. This next wave is more likely going to be seeing technologies that have historically been isolated to the consumer leisure space not just start to test the water in industry and enterprise proper, but really become mainstream to proliferate and to drive the growth of these categories.

Marc Petit:

So you played an active role. Actually, Zuckerberg interviewed you. You did not interview Zuckerberg at the the launch of Meta. In hindsight, what do you think has been the impact of this announcement? I mean, it was a big move, $10 billion of investment. So what's, six months after that, what do we, how do we look back at that event?

Matthew Ball:

Yeah. I'd say that there are three primary impacts. Some are good and some are bad. Number one is it has really helped to popularize and familiarize the entire world with this theme, and also legitimize it in the eyes of many. If you are a founder trying to build in the metaverse, having greater awareness and validation, whether that's through advertisers, investors, or consumers, is good. And so we've really seen a wellspring inactivity that is all positive. The downside may be the way in which for that same reason Facebook's specific vision of the future, which is largely conflated with virtual reality devices or immersive VR headsets, has kind of seeped in and started to characterize it. I would certainly not say that VR is the metaverse. I wouldn't say that it's even strictly required, and over the next decade I would be shocked if it were anywhere near the most common device or interface for accessing the metaverse, and as I just mentioned, we're probably some ways from it even being mainstream viable. And so that has, to some extent, mixed up narratives in a way that is often counterproductive.

Matthew Ball:

But the third is the way in which it has really just set off this global race now, and we can see this in a few different ways. Amazon, in the weeks that came thereafter, began rewriting their job descriptions to focus on the metaverse. In fact, they often started to emphasize specifically that they were building experiences in unreal. Google has reportedly reorganized their head of VR and special products. Everything in 3D now reports to the CEO of Google, and the entire kind of incubation arm sits underneath him.Tencent has doubled its initiatives in the space. And so now we're seeing every company that has spent decades wondering when it's ready to start building now sees it to be the present day.

Marc Petit:

So it's interesting that you call out the conflation with VR, because for me, Facebook conflicts more with this advertising model, and that economical model of putting out free products, network effects, and pushing to people advertising that may actually, or may not desire. And so do you think we were ... Well, especially if I speak for Epic, we had ambition that the new generation of platforms would go off to a different model, more commerce. Would take a precedence over that passive advertising model. So what do you think we are in that conversation? And is Facebook, we'll see would be successful and try to further their data-driven culture into the metaverse?

Matthew Ball:

I think there are a few different ways in which we can break this down. First and foremost, Facebook is an incredibly concentrated business, and to their detriment. Right? Almost all of their revenues come from advertisements. They don't really have a platform business. They don't really have a hardware business. They're purely reliant upon the advertising economy. They're obviously operating one of the most successful businesses in the modern era, if not history, and yet they still have exposures. And diversification is generally seen as a good thing. I do think that no matter how important advertising is in the future, they definitely aspire for more diverse and less healthier business.

Matthew Ball:

The second thing is to recognize that we actually can size the advertising economy pretty effectively. It's roughly 1% of GDP. Ever since the 1920s, with the exception of the Second World War, advertising as a share of GDP has always been between 0.9 and 1.3% of the economy. It's shrunk considerably with digital ad efficiency and targeting. It's actually one of the most remarkable effects, is after decades, literally near a century of oscillating between those two bands, we've seen not just stability for a decade, but we've actually trimmed it down to about nine tenths of a percent. That suggests the role of advertising in our society. And it also shows that for all of the ills of targeting, it actually has shrunk advertising share of the economy overall.

Matthew Ball:

In gaming, of course as we know, advertising is not commensurately revenue generative compared to other mediums. It's more than half in TV, in print it's roughly half. It's nowhere near that, especially in 3D-rendered, rich, virtual worlds. With that in mind, I tend to believe that we're going to see more advertising in 3D worlds in the years to come. Pretty much every medium that has resisted it eventually gets ads for the very reason in which we don't have them today, which is scarcity.

Matthew Ball:

Think of streaming video. Most streaming video services don't have ads. HBO added them, Disney+ is adding them. Netflix has spent a decade and a half saying they're not going to have them, and now they're saying never say never. It's specifically because they don't have them that the economic incentives to have them become so overwhelming. But overall more functionality tends to lead to more sophisticated methods of monetization. And we tend to say that that is user spending rather than advertisements. Television started as a relatively clumsy medium. It was broadcast undifferentiated. Didn't matter what you were watching, where you lived, what you liked, everyone was getting the same broadcast to their home at the same time. And it was all commercials. We moved to user-paid model, and now pure play subscription model.

Matthew Ball:

So if the internet economy over the last 15 years on the consumer side has been primarily ad-driven, I would expect that if the metaverse elevates that to a 3D version of the internet, with more things to sell, frankly, we will start to diversify the entirety of the internet to a healthier model, which tends to be the things people want to buy, rather than the ads that are shoved in their face.

Marc Petit:

It's interesting that the data says that the advertising portion shrinks in the GDP. From a user experts perspective, it doesn't feel like this at all. I don't feel I'm exposed to less advertising, so the efficiency must be there. So Patrick, up to you.

Patrick Cozzi:

Yeah. Let's switch gears a bit and dive a little more into the platforms. So Matthew, we're seeing two areas where platforms are coming from. One is all of the game platforms like Fortnite, Roblox, Minecraft, or PUBG emerging into the metaverse, but then we're also seeing ground up platforms such as Meta Horizon, or Decentraland. Wanted to see if you had any observations or more predictions here.

Matthew Ball:

Sure. So it's a fascinating time to think about this, because I kind of ... I often phrase the opportunity the following way; if you believe that the metaverse is a multi-trillion dollar opportunity, if you believe that it is a next-generation version of the internet, then you have to believe that billions of people will participate in the metaverse on a daily basis. If so, the point in time in which we're sitting right now, we're existing, is kind of like the ICQ era of online, or perhaps 2005. In 2008 even, Facebook had 150 million MAUs, monthly users. In 2009, they had 350 million monthly active users. A decade and a half later, not even, they had 3 billion MAUs, and 2 billion DAUs. The largest virtual platforms in the west are Roblox and Minecraft. They're sitting at roughly 150 to 250 million monthly active users, and about 30 to 50 million or 55 million daily active users.

Matthew Ball:

We're actually quite small. We have billions of users yet to adopt these services. And so what that means is the field is actually a lot more open than we tend to believe. The opportunity is mostly in front of us, not captured. Now we may see a version of the past in which the most dominant platforms like Facebook or YouTube, two of the leaders in the PC and early mobile social UGC era, continue to get larger. But there's also the opportunity for other companies to displace them. And certainly, we know that part of Facebook's attendance came from acquisitions of Instagram and WhatsApp, and had they not, it's quite likely that Facebook would've been supplanted. But even if today's leaders grow, the other lesson of the past decade and a half, is that new platforms have the opportunity to amass hundreds of millions or more in users.

Matthew Ball:

Snapchat was created in 2011. I think Twitch was started that same year. TikTok began in 2014, but then was really rebuilt in 2016. And of course, Fortnite now has a social graph of in excess of 500 million users with, I believe, 2.7 billion player connections came out in 2017. And so that is to say, there's a good reason why all of these platforms are being designed, are being launched, are being tested, and that's because most of the market is in front of them. And even if they don't become the leader, they have the opportunity to build hundreds of billions of dollars in value.

Matthew Ball:

What's interesting here is that, as you would expect, and as we saw with Instagram and TikTok, neither of which directly compete with Facebook in terms of functions, but do in the social networking space, these new platforms are taking different technical approaches. Horizon Worlds at Meta is clearly organized around VR. Rec Room and VRChat are as well. Niantic is clearly working at this from an AR and location-based perspective. And then when you take a look at the Decentraland, or in particular Sandbox, they're both built on Unity, but they're trying to jumpstart their economies around what is today very lucrative, albeit speculative, NFT system. And so we're seeing different versions of which audiences you start with, which technologies you prioritize, which experiences you look for, the structures of that economy. And time will prove which of those angles is actually the best.

Marc Petit:

So does that explain the frenzy that we are seeing from VC actually trying to invest massively in games that we haven't seen. Visual effects company like DNEG go public very successfully, or Technicolor are spinning off their creative services arms. So is this the race to fuel those in the hope of striking a winning platform or IP? How do you interpret that frenzy of investment in our space? And the consolidation, by the way, that we're seeing with the Microsoft acquisition of activation and all of that stuff.

Matthew Ball:

Right. And I'd say there's a fascinating element of this, which is let's go back 20 years. Disruption theory, innovation, these were not actually terms that many people spoke about. The average person never thought about them. It was never on the cover of Time magazine, and boardrooms weren't obsessed with it. We have all gotten smarter over the past 20 years in particular. And we've generally learned that each of these successive platform waves of which I would say that there have been three thus far, the mainframe era, which ran from the '50s to the '70s, followed by the PC era, which was coincident with the rise of TCP/IP or the internet protocol suite, and then the last 15 years of mobile and cloud. We've learned that they tend to come faster than we expect. Each one has been more valuable than the last. And they tend to disrupt nearly every category, even the stodgiest and most protected, whether that's finance or telecommunications.

Matthew Ball:

And so I think every single company is now more observant of this. And that's why we see the faster and more aggressive rush to the metaverse than we've ever seen before. Facebook almost missed mobile were it not for Instagram, and many companies like Microsoft or Blackberry were displaced in that chaos. No one wants that to happen again, and so they're being aggressive, they're being early, and they're willing to make the bets.

Matthew Ball:

I heard recently that last year, as many know, the blockchain space received 34 billion in venture funding, four and a half went to gaming. But more importantly, an estimated 120 to 200 billion is currently earmarked or raised for future investment that will probably happen in 2022, 2023, and to a lesser extent, 2024 based on fund cycles. This is everyone believing that this nebulous concept of the metaverse of Web3 of blockchain is a fundamentally successor version of the internet. And therefore the $25 trillion digital economy is increasingly up for grabs. And so that's why we see such swings in stock prices. That's why we see these types of companies that have never gained much interest go into the spotlight, and that's why we see such aggressive investment.

Matthew Ball:

But I think the interesting thing that reflects both of what you're doing is this growing appreciation that 3D rendering, especially real-time 3D rendering is the skillset of the future, which for 25 years has not been considered particularly relevant outside of consumer leisure. I use this quote all the time, I think his name's Don Evans. He was the US commerce secretary, and in 2001 he made this point that today's supercomputer is tomorrow's PlayStation. We talk about the fact that the metaverse is here in 2022, that Snow Crash coined the term in 1992. NVIDIA was founded nine after Snow Crash came out, coincidentally, but Jensen has been very clear that he built the company for 30 years for the era of graphics-based computing, believing it was needed to solve queries that general purpose computing could not.

Matthew Ball:

And for 30 years, the primary use-case for NVIDIA's technology was GPUs for video games. And all of a sudden we're recognizing, no, actually the entire world is eventually going to need the capabilities that GPUs can produce. And so I think we're just seeing this often diamond in the rough or uncovered gem situation, where all of a sudden we realize not only were these overlooked capabilities and companies, but there actually aren't many of them. I think many people are shocked to find out that NVIDIA is now the seventh largest company in the world. They're larger than Meta, they're larger than Tencent. It's unbelievable.

Marc Petit:

Absolutely. And you did a good job in one of your essay in highlighting how far a model like Roblox is from profitability. And we know Amazon took many years to get to profitability. Now they are public companies, so I guess we have good insight into their financial. How is underlying economical model doing? Are we seeing some improvement? Are we on the right trajectory? You kind of made the case that it would solve itself through scale. Is this still the thinking?

Matthew Ball:

Yes. And so for context, Roblox's P&L actually provides incredible transparency into what happens to the average $100 in user spending. We know that roughly 30% of it goes to trust and safety plus infrastructure costs. There's about 7% that goes to SG&A and marketing. There's 4% that goes to overhead. And some of those elements have exposure to cost savings with scale over time. SG&A should not scale commensurate with revenue in perpetuity, even trust and safety should see some improvements over time.

Matthew Ball:

Other areas will be more challenging. Infrastructure is unlikely to materially improve because the technology will improve. They'll get more powerful CPUs, they'll try to support higher concurrency. They'll try to do more cloud data streaming, which is going to increase their bandwidth costs. But what's fascinating to me is the two other areas. And actually, let me take a beat.

Matthew Ball:

We've seen few percentage points improvements in Roblox's profitability over the past year, not substantial, they're still losing roughly 26% on revenue. So basically their costs are 126% of revenue, but there are two primary buckets that are worth considering. Number one is R&D. They now spend more in a quarter on R&D than they had in revenue in the first quarter of 2020. And that's because they have maintained R&D as a percentage of revenue, even as their revenue has grown 450% since the pandemic.

Matthew Ball:

This year, they are likely to spend 600 or 700 million R&D. For context, Sony's entire video gaming R&D budget is about 1.2 billion, that spans network software and services, all of their individual studios, as well as the console itself. Unity is spending 300 million in R&D. I would assume that Epic's Unreal business is somewhere in the several hundred million.

Matthew Ball:

For further context the five year build for Red Dead Redemption 2 was about 250 million in production cost, probably another 100 million in marketing and publishing cost. The fact that Roblox's spending 600 million or 700 million in what they conceive to be sustainably is astonishing.

Matthew Ball:

And so some of that will eventually ease back. They can't continue to grow revenues in perpetuity while scaling R&D. But it's quite remarkable the extent to which they are productively investing. And I know you had the chief science officer talking about where that investment's going. I think episode 11 or so, talking about higher concurrency, natural language processing, real time translation, 3D spatial audio. And so a lot of this is healthy. Roblox is choosing to forego current profitability in order to build a better product over time. But the second issue, and this is arguably the biggest issue is roughly 30% of their revenue gets paid right away to the app stores.

Matthew Ball:

This is a situation that has two different consequences. Number one is it means that essentially provably, Apple and Google will forever make more money from Roblox than Roblox ever does. It also means that to the extent in which Roblox is confined in its ability to reinvest in its platform, which would be good for its developers and good for its consumers, that's constrained by 30% payment that they have to pay to a platform who's not involved in said exercises.

Matthew Ball:

The second is to recognize that developer payments are quite paltry. The average Roblox developer gets about 26% of all revenue spent through their system. But again, that's partly because Roblox themselves only get 70 cents on every 100 that come in the system. And so that 30 cents means Apple or Android are not only making more than the platform that is enabling this activity, they're also making more than the developers are. And even combined, the two categories are likely to get less. And so the biggest issue is one that Roblox can never scale out of.

Marc Petit:

Maybe. Let's see. Let's see if the needle could move on that topic. But not something I want to talk about, but you guys know what I'm thinking about. So Patrick, up to you.

Patrick Cozzi:

Let's talk about interoperability. So Matthew, as you know a big theme on this podcast is the openness within the metaverse and allowing lots of different software and vendors to work together. And you write about an expression, you call the idealism of interoperability. So I'd love to hear you talk about that.

Matthew Ball:

So interoperability is a really interesting topic. And one of my pet peeves is when people talk about interoperable assets. And of course, as we all know there's not really any such thing as interoperable assets because there's no such thing as a virtual asset, it doesn't really exist. What we're talking about here is data. And for data to be interoperable, so to speak, we're really talking consistencies in conventions, in code, other frameworks that allow us to exchange data, to understand that data, and then put that data to work. That is usually a business or a policy problem, because it is a decision that a given platform has to make and that they have to then act upon.

Matthew Ball:

I like to talk about the fact that it's actually the reverse of the real world. This may be an interesting way to think about it. Let's think about Nike and Adidas. If you want to wear your Nike shoes into an Adidas store, Adidas has to make an active decision to prohibit you from wearing Nike shoes. What does an active decision mean? Well, they have to have a policy. Then they have to invest in, basically, security guards. And then they have to actually do something to keep you from keeping those shoes in.

Matthew Ball:

The passive approach is just letting you come in with the competing shoes. Now we can all recognize that it would be absurd for Adidas to say no Nike shoes, but what's important is to recognize not making a decision means supporting or being compatible with foreign shoes. And that's because atoms are essentially right once, run anywhere. And no one makes the decision. They're preset, they're immutable. In the virtual world the reverse is true. If there's a given platform, let's say a standalone version of Nike, even if it's running on Unreal and Adidas is using Unreal, it's an active choice.

Matthew Ball:

They need to actively decide to incorporate and accept data, a virtual item in shoe from a third party. Then they need to build systems that can understand that data and then coherently deploy it. What does that mean? Well, Adidas would have to say, "I'm going to take the Nike shoe. I'm going to recognize it's a shoe." Because keep in mind, they could see the data and then say let's manifest it as an elephant.

Matthew Ball:

And then they have to say, "Let's apply its attributes the way that it's intended." Mark, I know you gave this brilliant interview with IEEE that I reference all the time where you talk about the fact that avatars are really tricky and 3D objects are tricky. It's relatively easy for a .jpeg to be converted to a .gif. It's relatively easy to go from PNG to JPEG because we're talking about a 2D image. But when you're talking about a virtual good, you have these other things, which is okay when you're bringing a Nike shoe, is the shoe the object or is the sole an object? Are the laces separate? Are the lace caps separate? Are all of these customizable? Do the shoe laces move with the wind and if they move, what's their weight?

Matthew Ball:

So we're talking about all of these different active decisions and it doesn't just R require investment and decision making. So they sometimes have cost. The more data you bring into a system, the more burdensome it is. Maybe allowing shoes with waving laces is actually burdening your render so that you can't do some of the other things you would want to do. And so this is a tricky topic. You started this by saying we weren't going to get too technical. So I'll try to back out of this. But so the idealism of interoperability is the hope that we can nevertheless find basically a bunch of buzz words to still support this. This is where people talk about frameworks of frameworks, systems of systems, data conventions, using machine learning to upscale and downscale different assets and data so that they're appropriate.

Matthew Ball:

So that if you don't need the shoes to have five different components, you can make the intelligent decision to treat them all as one. I'm optimistic that we will start to get to some of these. Why? Because the network effects of having ecosystems that can talk to one another are profound. We see this with the internet. From the 1970s to 1990s, we had the protocol wars. The consensus was not we would have one common inter networking standard. That was a radical idea.

Matthew Ball:

IBM, Comcast, Telefonica, China Mobile would all treat data and communication the same way. That was a radical belief. Even in the early 1990s, as the Department of Defense was supporting TCP/IP the Department of Commerce in the EU started supporting OSI, a competing networking standard. Eventually the networking standards converged, much like the global economy has eventually converged in globalization around things like the metric system, English, the US dollar, the intermodal shipping container.

Matthew Ball:

These conventions enable the exchange of information, English, or goods, or commerce. But what's key here is to understand this is not binary. We don't have interoperability or not in the internet today we have public and private networks. We have paywalls. We have private protocols that sit outside of TCP/IP, and we don't have everything exchangeable. In the global world today you can't bring $10,001 into the United States in cash. You can't bring a gun into a hospital. We have limitations.

Matthew Ball:

And so the most important thing to understand about interoperability is it's going to be inexhaustive, it's going to be incomplete, and it's often going to have a cost, either literal in terms of what you have to pay to bring something in or in function. Maybe you can bring your Nike shoe into Fortnite or Roblox, but perhaps you lose the five different elements that you can customize. Perhaps you can wear it, but the metadata that says you can run fast with it isn't brought with you. These are some of the considerations that we'll have, but again, note that's the same in the real world. You can bring your basketball into a museum, but you can't walk the floor with it. And you certainly can't dribble it.

Marc Petit:

Yeah. Makes sense. So interesting, I think interoperability we're focusing on making it half happen, but we know it will only happen in the context of policies and business models and economical decisions. So one of the other thing that we haven't actually really touched on, on the podcast so far, we're staying away from the topic is all of those decentralized technologies, and NFTs, and the blockchain and all of those new arguably universal technology to support new forms of economy.

Marc Petit:

We see a lot of scams, to be honest with it, we don't quite understand it. And we see a lot of scams, so we don't get anywhere close to it. So what is your... Let's venture into this territory with you? What is your take of the current option of all this Web3 stack and the promise of universal, on the ground commerce with NFTs and blockchain and all of that stuff.

Matthew Ball:

I kind of have three different approaches. Number one is I'm excited about some elements of the philosophy, which is to say it is pointing a very clear picture that users do value interoperability to some degree, and that various autonomous or independent organizations, game devs, platforms can use a technology that they don't own that does have some technical concessions in terms of functionality and exchange assets, creations, users, even currency with other platforms in a way that is feasible and that can be seen to rise the tide.

Matthew Ball:

This question we mentioned earlier of interoperability ultimately comes down to business practices and philosophies. And so having at least the groundswell we saw last year, the economic potential and the express wants of consumers is encouraging because it will lead to other parties saying, "We should give this a go. We should open up. Maybe we don't have to have indefinite control of everything we do in order to maximize profit."

Matthew Ball:

The second is to recognize that most of what I find interesting in crypto or blockchain is more the decentralized operation of infrastructure. We have companies like the Render Network in OTOY or Helium. And what they're essentially doing is providing trustless, permissionless, programmatic ways to share infrastructure like we share solar panels on top of our homes. If there's one thing that most people who focus on the metaverse can agree upon, it's that the requirements in terms of bandwidth and computing power are extraordinary and far beyond anything that we have available right now.

And so having some sort of decentralized system that can automatically compensate people for the provision of bandwidth or computing resources is a positive step forward for which blockchain is not an essential or required technology but does seem to be a successful version today. The third thing that I'd say is we need to recognize that these technologies may be decentralized in the specific instance, but most of the experiences that leverage them remain heavily centralized to the point in which you have a choke hold.

Matthew Ball:

So let's talk about any of these virtual world platforms or the blockchain-based games. This is a good way to get back to the point about data, which is, sure, the rights to your NFT may be decentralized, but the asset itself is still stored on a centralized server. But critically the game itself is still running on a centralized server, the code. A Unity game or an Unreal game is not running in a decentralized fashion. And it's probably physically impossible to have a highly rendered 3D virtual game running on decentralized infrastructure, certainly around the world but even potentially up and down the streets of your local neighborhood.

Matthew Ball:

And what that means is you may have a perfect receipt that says you own this digital NFT, but if a game wants to reject your NFT, they can. If they want to say, "Marc, you have a Pokemon, you have a Charizard with 100 HP damaging Fire Blast, but for meta reasons we've decided we have to nuff it down to 80, they can do that. And that's because you have the data that requires code to operate, and that code is operated in a centralized fashion.

Matthew Ball:

And that's just one of the many technical problems. There's also the question of your account system, the leader boards, matchmaking. All of these remain centralized. And so there's a very legitimate argument, which is to say, if anything is centralized, then nothing is decentralized because you've still got a choke hold. Regardless, this is where we get into the philosophy point that I kicked off in, which is we are operating in a world in which decentralized gains are still majority centralized, but the philosophies of these systems are requiring or seem to be requiring different or better behavior from those companies that are operating the centralized code. And so that may just be the better outcome out of a technology that today is not inexhaustive. But you're right beyond that. I mean, almost everything that we see is still scams and wash trading.

Patrick Cozzi:

So Matthew, I follow you closely on Twitter and I know you're working on a new book. Tell us about it.

Matthew Ball:

Well, so I've in theory mostly finished it, but from August until late January, I was working on a book. It'll be published by Norton. It's called The Metaverse: And How It Will Revolutionize Everything. It's on sale for pre-order at every retailer that you can think of. It will come out in early July. It's 350 pages, about 105,000 words, and it goes through three core topics. The first section is really defining this relatively inchoate and confusing topic while explaining its origination history, some of the lessons we can learn from 50 years of trying to build these experiences, and touching on some of the topics we just discussed, which is, why is gaming at the forefront? Why is this next generation internet starting from a leisure category whereas the last wave started from enterprise and government?

Matthew Ball:

The second section walks through all key technical issues. The hardware needs, compute needs, networking needs, the role or potential role of blockchain, virtual world platforms and so forth. And then the last section really details some of the societal elements. How might it change the world in which we live in? What is the role or non role of regulators? And then most importantly, what can we be certain about, but what can we not know until it arrives? And this is where I try to take the backseat to a cursive innovation.

Marc Petit:

That sounds fascinating. I mean, yeah, it will be a must read I think for most of us and probably for most of our audience. So July 19th, right? I think I read this on Amazon when I ordered my copies.

Matthew Ball:

Well, thank you. Yes, that's the goal. I'm hoping to push it up a week or two, but it's remarkable contending with modern supply chains post pandemic. It's quite something to try and look at, but I'm excited.

Marc Petit:

We're so looking forward to that, Matthew. So you've been following the podcast a little bit and moreso this conversation today. Is there any topic that's top of mind for you that you think we should cover and we haven't covered?

Matthew Ball:

It's a good question. I mean, one of the topics that I'm really interested in and we continue to talk about is the deeper and deeper we get into the digital era, the more we have to contend with a core technological problem, which is, the choice to deprecate old systems, to not be compatible with what came before. I get really interested about this from the metaverse perspective because we're talking about a persistent, almost parallel plane of existence. And so the nature of resistance means that we can't every hour or even year say all of that old stuff is gone. And so as a result, we have to have systems that ensure compatibility, up scaling, upgrading of what we already have to be compatible with what we want in the future, and yet there's that tension.

Matthew Ball:

We know that building from scratch allows you to have a cleaner tech stack sometimes. Obviously you end up overestimating how much you can replace and improve. But so we're going to go through this period as the metaverse starts to take off of trying to manage the legacy compatibility. And I was given this... Tim Sweeney has talked about one of the four big challenges being forward compatible code evolution and backwards compatible support.

Matthew Ball:

And I think of this story that Ebbe Altberg, the former and unfortunately late CEO of Second Life told me, where he talked about the fact that one day a few ago, Second Life upgraded their physics engine or the physics element of their engine and they found out a week or two later that there was a bug. And the bug was the specific business that was selling horses and horse feed found that the horses, post update, when they tried to bend over to eat the horse feed, would slip and they'd slip in perpetuity. And the result was anyone who purchased these horses ended up coming back to their game a few days later to find out that the horses had died. This bug literally caused all of the horses to die because you had dysfunction of a business model.

Matthew Ball:

Now, what was the challenge of this? Well, the challenge is Second Life is relatively decentralized. Second Life does not manage what's bought, what's sold. They don't administer the transactions and they don't observe them. And so they made an update that had millions of implications and they were almost all seamless, and yet there was one misfire. But then think about how do you solve that problem? What's the restitution? Who pays whom? How do you even know what the quantifiable damage is? How do you even understand the best way to do it? Do you roll back what happened to the horses? Do you give money?

Matthew Ball:

This is an isolated issue and it's not specifically about forward compatible code evolution. But if you say that we have this entire system of technologies, we have this vast interconnected economy that is predicated upon trust, I'm going to sell you a horse and you're going to use that horse. I'm going to buy a horse and my horse is going to live. And then you have to start contending with bugs, but most importantly, the need to update the physics of this world and then manage for the unintended and intended consequences. You actually get into moral issues.

Matthew Ball:

And so it'd be interesting just to see how people think about that issue. The only other one is getting the governmental perspective as to how do we start to regulate these things. If we do believe that this is a parallel point of existence, if we do believe that there's going to be multiple trillions of dollars here, then we actually need to do more than just modernize tech legislation on app stores, we have to probably recognize many of the rights that we take for granted in the physical world but are lacking entirely.

Patrick Cozzi:

So Matthew, to wrap up this episode, is there any person or organization you want to give a shout out to?

Matthew Ball:

That's a good question. So I'm helping to produce a title for a company called Genvid Technologies that's doing a mass scale interactive experience for the Walking Dead that comes out in July. It's going to support via cloud streaming upwards of a million concurrent viewers that will be able to place their own avatars into this live version of the Walking Dead. Robert Kirkman, who created the franchise, is creating his first characters in that universe in more than a decade. And so I'm really excited to see this experience come out because you are going to see for the first time ever the ability to actually place yourself or your creations into canonical Walking Dead content, and then have it play out in real time for millions who are also helping to shape that narrative.

Marc Petit:

Yeah. No, Genvid is a very interesting company with some really innovative and smart technology. So can't wait to see that either as well. So Matthew, thank you so much. Thank you for being there with us, the king of this new season. You're kind of a little bit of the godfather of this endeavor. So thank you so much. We want to thank our audience again. Keep on letting us know what you think about the podcast, the themes you guys want to hear about. And again, Matthew, thank you so much and thank you everybody. We'll be back for a new season.

Matthew Ball:

Thank you. See you guys soon.

Patrick Cozzi:

Thanks everybody. Thanks Matthew.