High Fidelity recently unveiled Avatar Island, a virtual reality domain where users can buy and sell items based on a blockchain service. Philip Rosedale, CEO of High Fidelity, sees it as the latest evolution of virtual worlds and virtual life.
While at Linden Lab, Rosedale created Second Life, one of the earliest and most successful virtual worlds, in 1999. Now he has a new startup in High Fidelity, which is a contender to build the same kind of virtual society in VR. But at his new company, Rosedale is trying to modernize the commerce system and open it up so it can be more empowering for both users and creators.
To do that, High Fidelity has built its Avatar Island commerce system on a blockchain, similar to cryptocurrencies such as Bitcoin and Ethereum. The ledger of transactions will exist independent of High Fidelity, providing unalterable, irrefutable proof of ownership. The system is in its beta stage now. It has a user interface for a wallet and marketplace. On the backend, High Fidelity is creating the first server for the blockchain, and it will eventually be one of many servers that store money and information about who owns what.
Under this design, High Fidelity the company could disappear and the blockchain would live on. Other virtual worlds could use it. It could even bridge the gap between purchases in the virtual world and the real one. With Avatar Island, users can now purchase more than 300 items for their avatars. Those items were created by digital artists around the world who will make money from the purchases. And if users want to take their belongings and avatars to another world, they will be able to do so.
In the future, item ownership recorded on High Fidelity’s blockchain could also be used to translate purchases between the virtual and real worlds. An asset’s PoP exists on the DAR blockchain independent of High Fidelity. An asset’s chain of ownership — its sale and resale — lies entirely in the hands of the object’s owner.
Here’s an edited transcript of our interview.
Phil Rosedale: Second Life still has a rich economy. That’s one of the pieces we’ve recognized we needed before people would start building, a focused system so people could buy and sell things with each other, and also a system to properly recognize ownership in a virtual world. In a place like Second Life, it’s relatively simple, because the game is under the control of a home server. The game company decides who owns what and keeps track of that.
But from the beginning, our vision for VR at High Fidelity has been — the nature of the way people are going to want to use VR and head-mounted displays in general for meetings and business and all these other things would mandate everybody running their own servers. When you look at something like the Second Life marketplace, which has something like 3.5 million items in it and $700 million in transactions — when you play that forward and say, “How does that work in a generalized VR sense?” you need something that spans all those servers, that allows people to recognize digital assets and make transactions. Just like a website, you don’t really trust any individual server.
We’ve been thinking about this and working on it for the last couple of quarters. As you can imagine, we first started looking at blockchain technology as the way to do. What we’re putting up in beta form next week, so everyone can start playing with it and using the UI and requesting money at the start, is a set of enhancements to High Fidelity that are basically — a UI for a wallet, for buying things in the marketplace and looking at what you own, and then on the back end that information is being put into a blockchain. We’re standing up the first server, and there will be many servers run by different organizations and individuals in a manner similar to the way things like Bitcoin work today.
That blockchain is going to store money. It will also store information about who owns what. The interesting thing about that is that — what we’re building, we envision it as a generalized digital asset registry. When you have something to sell in a virtual world — you could be Herman Miller wanting to sell official Herman Miller chairs, or someone who made an article of clothing they want to sell for avatars — the way we’ve built this, it’s similar to the patent and trademark office. You make a submission of something you want to register, and we put that up in public. We let some time go by, so that people can look at it in the same way you have a public period of inspection, and then once that period is over, we write a certificate into the public blockchain that says, “This item is now available for sale.”
Then, when somebody buys that item, they become owner in the same way you become owner of currency on a blockchain. Once you do that, you can then sell that digital good to anybody else. We don’t even need to be involved. It’s a completely public blockchain transaction. And then there is UI in High Fidelity that lets you do cool things, like if you walk up to someone and they’re wearing a hat, you can hover your lasers in the world at it and you can bring up the certificate, the proof of purchase on that item. This is all going into beta next Tuesday.
VentureBeat: How long were you guys thinking about blockchain as a solution for this? Did it come up a long time ago?
Rosedale: We started working on it about six months ago, formally, because it was at that point that we started working on how we were actually going to enable commerce overall. I’ve been looking at blockchain, Bitcoin and Ethereum and general blockchain tech, since Second Life, because obviously the Linden Dollar was in some ways the first digital currency. Not a decentralized cryptocurrency, but it was and is certainly the big digital goods currency.
About the time that Bitcoin came out, I was well aware of it, and I was also independently designing a kind of generic — I was thinking a lot, during my day to day duties at Second Life, about how to generalize what we had done at Second Life to do something decentralized for currency. I didn’t have all the ideas that Nakamoto had in the Bitcoin paper, but I’d started forming some ideas about a distributed ledger, basically, and how one might keep everybody maintaining the same ledger.
It’s been pretty natural. I’ve followed along with all of this technology. But the last couple of quarters we’ve finally dug in and started building the back end of this thing we’re putting out. It’s a lot of work. The blockchain technology is incredibly interesting and important, but it’s really barely working yet.
In particular — and this is an interesting element of the story, the way that public cryptocurrencies, this is Bitcoin and Ethereum, the two big ones, work today — because of the way consensus is reached, by doing a puzzle basically, a cryptographic lottery game in which somebody wins after a few minutes, the way those systems works puts a strict limit on how long it takes to confirm a transaction and how many transactions can be done per unit of time. Finally, the cost of doing a transaction is set in a kind of uncontrolled way by these big public networks. There are several problems.
VB: Why not use Bitcoin or Ethereum?
Rosedale: A way to achieve our decentralization goal would be to say, “We’ll just store this certificate information and a currency for High Fidelity on one of these public blockchains.” In fact, we could do that, technically, but a couple of big problems come up.
The thing about a public blockchain we wanted to take advantage of was that the most durable representation of a digital good is one in which there’s no central company that could accidentally or intentionally delete it, if you already own it. This is also true for money, of course, which is what’s compelling about blockchain. It’s a kind of money, or a database record, that can’t be altered by a central agency. It felt, to me, that — again, in keeping with this idea that with High Fidelity we’re trying to build open source, publicly usable software that millions will be using — the financial system and the recognition system as to who owns what would logically be best done as this type of a public blockchain.
The problem with using the existing blockchain is that the settlement time is too long. That won’t work, because if you imagine a transaction in the virtual world, it might be between two individuals face to face. Or you’re buying something from an online marketplace, but you’re standing there waiting for your new car to show up in front of you so you can get in and drive it. Because of the way Bitcoin and Ethereum do their settlements, they impose a long delay and a high fee for each transaction. I don’t know what Bitcoin is over the last couple of days, but it’s probably about $10 per transaction. It makes sense if you’re moving large amounts of money around. But as we know from Second Life, digital goods are typically a couple of dollars. You would have to have a transaction fee — it doesn’t have to be zero, but it would need to be a couple of cents.
For that reason, we had to dig into the problem of, Do we use a different blockchain, or do we need to build our own? For the time being, we need to build our own, because there aren’t any blockchains that are up right now that offer higher transaction speeds and lower fees. In the future, as systems like those come online, we may shift over to using those, because the actual blockchain, on the very back end itself, we want it to be as widely used and as generalized as possible. That’s the promise. We want as many people backing up the data, if you will, as possible. But for right now we’re building a blockchain we can deploy and have others deploy as a modification of existing Bitcoin software, modified in such a way that it can have very rapid settlement times and very low transaction fees.
The other complicated thing about a virtual world, and we saw this in Second Life, is that if you want people to use a currency to do things like buy clothes for their avatar, the currency can’t be wildly fluctuating or going up in value. Obviously, if that’s the case, people will just hold and accumulate the currency rather than using it for transactions. That’s what’s happening with Bitcoin and Ethereum in a very public way right now. We couldn’t use Ethereum as a currency because no one would buy clothes for their avatar. Everyone would just hoard it.
For those reasons, we’re creating a different system. We will probably tokenize — as you may have heard before, you can connect one blockchain to another in such a way that you can move a token from our very fast blockchain onto the Ethereum blockchain. That’s not a mechanism we’re deploying next week, but we’ll almost certainly be doing it in the coming couple of months.
That will allow people to do several things: for example, to move the currency they earn from selling things in High Fidelity over to a more heavily used public blockchain like Ethereum, for the purpose of selling it on an exchange market to turn it back into U.S. dollars. The other reason you could imagine as far as why we would choose to use a blockchain like this — if you can make a blockchain currency convertible to other blockchain currencies, then it provides a convenient mechanism for users of virtual worlds to take their money in and out of various other currencies, and we don’t have to do that work. We don’t want to do that. We want to focus on the actual VR software. That’s another reason why we’re taking this particular approach.
VB: That seems like an interesting decision you would have to make. In some ways, if people are buying currency and they stay inside the world, they do trade a lot of that currency and make a lot of transactions. But if there’s a way to cash it out, they may try to do that, and then go to another game that also uses their own blockchain currency. Are there interconnectedness issues there, some risk that your people might leave you?
Rosedale: That’s a good way of putting it. We’re willing to take on that risk. After studying this a lot, we feel that maintaining the convenience with which you can move cryptocurrencies around and change them into each other has so many more benefits, both to us and to our end users, that they vastly outweigh any attempt we might make to keep money in our system or encourage people to spend it internally versus taking it out and turning it into Bitcoin or something else. We want to enable those transfers in every way that we can.
We’re already in conversations with the various currency exchanges and other services such as wallet providers, working through how to recognize the currency that we’re standing up. It’ll be another token, basically, and another — the most common technical comparison would be to, say, AltCoin, because that’s the expression for the blockchain that we’re actually putting up ourselves. Technically we call it an AltCoin. We’ll also tokenize the currency and make it exist, most likely, on the Ethereum blockchain.
VB: I’ve talked to Tim Sweeney at Epic Games a few times about this. He was saying that if you’re going to build an open Metaverse, which he says is possible in the next few years, you want to do it on blockchain, because it can force each of these individual worlds to be more open through this interconnectedness.
Rosedale: Right. I think specifically, the two simplest things that need to be open to make the Metaverse the most useful it can be for humans are identity — that is, Dean can come to my server for a meeting and prove that he’s Dean — you should be able to do that in a way that’s blockchain-based, never violates your privacy unless you want to — you show me your card at the moment you come to the meeting — and works across multiple virtual worlds. The other thing that needs to move across virtual worlds is digital assets, like the clothes you wear.
That’s exactly what we’re trying to do. We’ve written an open source client and server for High Fidelity. People are putting up those servers. We’re still early in our beta process. We have a lot of work to do, the least of which is this whole commerce system I’m telling you about, but we have about 125 domains up. Those all need to have identity and assets that can be moved around between them. Moreover, we all have to find some sort of standards for the Metaverse, or for virtual world environments — I should be able to take my digital Ferrari and drive it around in Tim’s space. This is the first step toward that, in that we’re trying to establish a currency and a digital asset registry so we can put these certificates on the blockchain.
The other thing that’s compelling about this, I believe it’s an immediately useful use case for blockchain, and in particular for the idea of representing assets on a blockchain. There are tons of startups doing very interesting things, aspirationally, on a blockchain. But a lot of them are quite early. There’s not necessarily a ton of utility yet. But with virtual items there’s a tremendous need for interoperability that’s going to emerge immediately as VR takes off.
For example, right now you have all these different things like Facebook Spaces and Steam’s Home and the Oculus Home. I should be able to take a grandfather clock and put it in any one of those spaces. Moreover, if there’s going to be a vibrant economy around that, I need to also be able to carry along with that digital grandfather clock a proof of purchase of some kind, a certificate that says it’s official. That’s going to be what motivates me to actually buy the thing rather than just use a hack or make a digital knockoff.
VB: What’s your process going to look like for this, your road map?
Rosedale: We’ve done a lot of work to get this nice user experience up and running and into beta next week. The first process will be to grant some initial money in the cryptocurrency to our alpha and beta users. This will allow them to begin registering assets, as I described, and selling them to each other.
Next week — we have several hundred things from existing creators in our marketplace already, and we’ll also be putting up an avatar store. It’s a big shopping center on a High Fidelity server that has a bunch of attachments and accessories for avatars. On Tuesday you’ll be able to go shopping and get glasses and hats and backpacks and things like that to put on your High Fidelity avatar, and when you make that purchase, you’ll be going through what I described. You’re purchasing an official object. The certificate is put on the blockchain for you. Anyone who runs into you in High Fidelity can hover on that thing, pop up the certificate, and see where you bought it. They can also buy one of their own, which is obviously one of the benefits of digital goods in a virtual world.
In terms of road map, the next thing we’ll be doing is making the connection to other cryptocurrency systems, which will allow things like currency exchanges, so you can take money in our currency and turn it back into real money. We’ll be doing that as soon as we can, in another month or two after this beta goes up. We’re also in talks with other virtual world providers or interest groups about having them run additional nodes, as we say, in this blockchain we’re putting up, so we can begin the process of decentralizing the back end servers for the blockchain network.
There’s all kinds of littler stuff in the first quarter: flexible clothing for avatars, more things like person-to-person payments, walking up to someone and handing them currency. You can’t actually do that next Tuesday, but you’ll be able to soon. Next week we may also have a blockchain explorer up. If not, it’ll be imminent. That’ll be like a webpage, kind of like what you see with Bitcoin and Ethereum, that lets you inspect the public blockchain itself and see where everything is from the vantage point of a webpage rather than needing to go into the world.
We’re going to aggressively engage in a process of trying to get other virtual world systems to use this systems. Again, we’re very much of the opinion that if we get more services to use it — if we get another social VR service using it, and we’re talking to a couple of them — it’ll be a benefit to all of us. We’re willing to undertake the risk. Maybe our customers will spend their time hanging out in other systems where we can’t monetize them, but we’re willing to take that risk to get a larger Metaverse network set up.
VB: Is there a point where you would expect to try to make alliances with the likes of IBM or the accounting firms, the people who are doing large blockchain commerce initiatives?
Rosedale: I don’t know if you remember what a great participant in Second Life IBM was. They continue to do stuff there, I believe. I think some of the forward-looking infrastructure companies, like IBM or folks like Amazon, are the logical kinds of partners for pushing the agenda of interoperable — another great company is Google. They put up their Poly API this morning, and we were mentioned in that release, because we’ve been working with them to allow you to drag and drop content from Google’s Poly site.
I don’t know if you’ve seen that yet, the Google Blocks and Tilt Brush stuff in VR. You can drop those things into High Fidelity and have them just show up. Right now they don’t have any rules and regulations around reuse. They just have a big everything’s-free site. That would be another example of the kind of logical partner that could take advantage of such a system. It would be amazing to have people making things for Google Blocks, and they would be able to put a price tag on them and have them certified, no matter what virtual world they were in, in a similar fashion. Other things would be content sites like Turbosquid and things like that, but we’re moving as quickly as we can in terms of reaching out and talking to people.
VB: It seems like you can use this and turn it into a much bigger idea than just “Here’s one more virtual world.”
Rosedale: One of the specific things we like in that sense: Imagine a luxury item that’s interestingly unique in the real world. Jimmy Choo shoes or something like that. One of the other things we’d love to do — we haven’t started reaching out on this yet because we’re a small company and it’s further down the road, but we think it’s a possibility — would be to say, if you buy those Jimmy Choo shoes in the real world, maybe you’d get a certificate, issued automatically, which enables you to wear those same shoes inside High Fidelity or anybody else that’s participating in this system. That would be pretty cool.
I was talking to someone about this and I said, “Somewhere in New York is a person who owns more Jimmy Choo shoes than anywhere else in the world.” It’s an interesting property of public blockchain that you could give that person the ability to walk into Sak’s Fifth Avenue or whatever and have that little fact be something that they could know, or share with them. I’ve been struck by how many interesting opportunities there are.
I think the challenge is a chicken-and-egg problem. With respect to hard goods in the real world, labeling them all and putting them in the blockchain — it’s not that strong a use case yet. It’s not a very interesting thing to do. You have to have enormous adoption mass before anyone cares. But on the other hand, taking, say, real-world items, even a few cool real-world items, and having you get the virtual version along with them, actually seems like a pretty good idea. It’s the kind of thing that makes sense for blockchain.
VB: Do you think there’s any risk of — what happens if Bitcoin goes back down to a dollar each, some kind of crash?
Rosedale: That was one thing I hadn’t touched on yet, but the other thing we’re trying to do is create a stable exchange rate. Second Life has had a very stable exchange rate, about 265 Linden Dollars to the U.S. dollar, for the last decade now. It did that during an S-curve period of very slow growth, very intense growth, and then very slow growth again. Over that whole period we had a very stable exchange rate against the dollar and the Euro, and we still do. You can go on the Lindex and see what the exchange rate is. I guarantee it’s right around 260.
Bitcoin, on the other hand, is ridiculously unstable and unsuitable for commerce at the time, whether it’s $10,000 or $1,000 or $1. It just fluctuates too quickly. What we’re going to do there, people haven’t started doing this with blockchain yet, but there’s a way to do it and a couple of great papers I could share. There is an algorithmic way, using smart contracts, to manage the money supply of a currency that makes it behave the same way that Second Life’s did.
The question is, how do we maintain a steady exchange rate given that Second Life was totally open? It’s a market-based exchange rate. You can set your own buy and sell prices. There’s a real foreign exchange market. The price of the Linden Dollar in dollars is floating. But the way we kept it stable was we openly and transparently created new currency and sold it on the open market with the goal, as we explained to our end users, of holding the exchange rate roughly constant. When the exchange rate would start to drop, meaning Linden Dollars became more expensive, our trading guy would create new Linden Dollars and sell them on the open market.
The interesting question is, if you want to do something like that, how do you do it if you’re using a cryptocurrency? Smart people have been thinking about this and we’re going to use the approach they’ve been converging on, which is that you can make a smart contract that monitors the exchange rate of the currency. That currency will trade against the dollar, or you could measure it against a basket of currencies or whatever. It’s what’s called using an oracle in the cryptocurrency world, meaning that your smart contract looks at a website to get a number, and when the currency becomes more expensive, the algorithm automatically disburses a bunch of new money. It mints new currency.
This is being done today with smart contracts on Ethereum. The capabilities are all there. Some of these different ICOs have been related to attempts at doing this. There are different ways of managing money supplies. But the basic idea is that — in High Fidelity right now we have basically zero people buying and selling digital assets. In a few years we’ll have many millions of people, or hopefully even hundreds of millions of people, engaging in VR digital asset transactions. The growth of that economy you manage using the process I just described. Rather than having there be 22 million Bitcoins ever, you have a smart contract that looks for the deflation rate and generates new currency and distributes it.
The way it’s distributed is a longer story. You basically create a second token on a public blockchain, and that token — whoever holds those tokens receives the newly created currency. It’s almost like a dividend from a stock. That’s something else we’re going to do next. We don’t need to do it right now because it’s easy to manage the currency at the outset. But that’s the strategy we’re going to use in the long term.
To your question about ICOs, that may be something in the future that we sell, those tokens that serve as the recipients of those dividends. That’s a piece that we haven’t worked out, but that’s directionally what we’re going to do. What we know is that the outcome of that will be that the exchange rate can be managed to be relatively constant. That will make the currency, the HFC, completely usable for day to day transactions.
VB: It seems like you’ve become a Secretary of the Treasury for High Fidelity.
Rosedale: Well, that’s what I was for Second Life. A friend of mine who’s a big Bitcoin guy, they called him Alan Greenspan in Second Life because he was so obsessed with virtual economies. He went on to become one of the big Bitcoin entrepreneurs. The interesting thing here, though, is to do it using an algorithmic approach, so that we don’t have to have a human-managed system. That would be a centralization risk, and one of the things we’re trying to do with High Fidelity and this broader ecosystem — we don’t want anybody to be able to convince me to do something stupid.