Chip Morningstar interview: How the metaverse started with Habitat

The metaverse is all the rage today, and its lineage goes way back in gaming. The universe of connected virtual worlds might well become a reality worth trillions of dollars, and it might be good to think about where it all came from.

The ideas were fleshed out in novels such as Snow Crash and Ready Player One. But games have played a big role in taking the idea of the metaverse from the realm of science fiction to everyday entertainment as well. It’s easy to think back to the influence of massively multiplayer online games like World of Warcraft or virtual worlds like Second Life.

But those worlds owe many of their features to Habitat, an online world released by Lucasfilm in 1986. If you can remember Habitat, then you’re really an OG. I came across a collection of materials from Habitat in 2020 when I visited the Museum of Art and Digital Entertainment (MADE) in Oakland, California. Habitat was built by game developers Randy Farmer, Janet Hunter, Aric Wilmunder, and Chip Morningstar.

It was the first real effort to build a large-scale commercial virtual community. Habitat was a graphical MUD, or multi-user dungeon, and was the forerunner of modern online communities. It had a short life on the Quantum Link online service for the Commodore 64 computer. Farmer and Morningstar were recognized for their efforts with a First Penguin Award at the 2001 Game Developers Choice Awards for their innovations with Habitat.

“Habitat continues to be the single most significant example of MMO history preservation at an institutional level,” said Alex Handy, former executive director of the MADE, which is open again in a new location after the pandemic prompted a move.

He added, “Habitat taught Chip and Randy a ton of things about online social interactions in virtual worlds. It taught them about distributed programming, event-driven programming, and object-oriented programming. Habitat laid out the model that has become the basic lesson of online social media: people can be really mean, and if they can break things, they will always do so.”

I caught up with Morningstar to talk about Habitat, the metaverse, and blockchain technology. Morningstar is a software engineer at Agoric, which is a platform using Javascript-based smart contracts to rapidly build deploy decentralized finance apps.

And he is still keeping up with all of the modern advances in MMOs, the ethics of the metaverse, humanism and technology, and blockchain games. We had an interesting conversation that spanned decades of technology.

Here’s an edited transcript of our interview.

GamesBeat: What are you doing these days?

Chip Morningstar: I’m working at Agoric on the software for the Agoric kernel, which is the low level piece that’s at the heart of the distributed asynchronous computation framework in the Agoric smart contract pipeline. I’m a couple of layers of abstraction away from the thing which is Agoric’s fundamental business, smart contracts. But it’s a thing which will enable mutually suspicious software agents to interact with each other and engage in contractual interactions of various kinds. Which has all kinds of lovely distributed computation problems at the heart of it, and that’s what I’m working on.

GamesBeat: It sounds like you’re a believer in blockchain technology.

Morningstar: Yeah, although Agoric is interesting. There are a lot of historical blockchain skeptics among the Agoric founders. The blockchain to us is not a religion. It’s a tool. It’s a means to an end. In our case the end is being able to have reliable, trustworthy computation that doesn’t require you to have an intermediary who is a trusted third party when you have parties contracting with each other. Rather, you get the effect of having a trusted third party by an emergent process from the blockchain. We’re using blockchain as our trusted computational engine for doing contracts.

GamesBeat: Making the blockchain more secure than it otherwise might be?

Morningstar: Well, I hope so. We have a fairly strong commitment to security above and beyond what historically the computer security business has concerned itself with. One of the interesting things about blockchain–one of our people, Mark Miller, referred to it as, “Imagine you have a billion-dollar bug bounty.” When you have a situation where somebody can make off with literally tens or hundreds of millions of dollars through an exploit in a matter of seconds with no recourse, suddenly you get to a place where people take security seriously in a way that they previously hadn’t.

One of the truisms in the security business has always been that nobody is willing to pay for security. In this case, you have some people who are willing to pay for security, and we’ve been very enthusiastic about an approach called object capabilities. I sometimes describe myself as a member of the dissenting faction in the computer security community, which is to say I like to think of the capabilities of security like the germ theory of disease in the latter half of the 19th century. A few people had figured things out and were saying that doctors should wash their hands. All the doctors said, “What do you mean?” I feel like we’ve been telling people to wash their hands. Capabilities are our approach to doing that. We hope that it’s an approach that will gain increasing traction as we go along, as it proves its value in actual adversarial environments.

GamesBeat: It feels like, with quantum computing coming along, the blockchain will need all the help it can get.

Morningstar: I don’t know. There are lots of post-quantum computing cryptography things that people have been working on. I’m not a cryptographer myself. I have lots of friends who are. I’m reasonably confident that they have lots of tricks up their sleeves still, so that even though some of the current cryptographic techniques might fall to quantum computation, there are other things that will not. There’s a potential for a transient hiccup, but I don’t see that as a long-term threat.

GamesBeat: It looks like you wound up at what could be a less controversial blockchain company. Something that could feel like it’s very useful for the future of the metaverse and virtual worlds.

Morningstar: I’ve been a virtual world guy since way back. You mentioned Habitat. I’m obviously familiar with Habitat, which not everybody is. I’ve found over the years that virtual worlds have had a tendency to be–the way I put it is they’re more interesting to their developers than their users, because of all the cool technical challenges that are involved. For something like metaverse, the utility of that to the mass market is yet to be demonstrated. It’s not that I don’t think there’s potentially some serious utility out there, but it’s still an open question.

Nevertheless it’s the case that building a metaverse or a virtual world is a great way to make a lot of the engineering problems and the security problems and the scaling problems and all of the technology–it makes a lot of those problems very concrete and very visceral. It gives you a way to visualize what you’re doing and how to think about it, in a way that’s easier than if you’re just thinking in terms of abstract protocols of some kind.

GamesBeat: I don’t know if it feels like the atomic bomb or something, where the metaverse is a guiding light for a lot of technologies whether people want it or not. It will advance technology.

Morningstar: A lot of it is that people have read things like Snow Crash or Ready Player One and have been very entranced by the “Wouldn’t that be cool?” aspect, without necessarily stopping to ask what you would do with it. I’m a bit of a skeptic. Despite being arguably a pioneer in the field, it is not obvious to me that there’s a use for this. I’d love there to be a use for it. I certainly am very open to somebody figuring out what the recipe is. With a lot of things, they wander around for a long time until somebody stumbles across the killer app, and then the killer app pulls everybody’s bacon out of the fire and allows it to go on to become great and successful. Nobody has the killer app for the metaverse.

I’m always a little nervous about the whole killer app idea, because it’s hoping that some miracle will come along and save you. It doesn’t seem like a good business plan to me. But nevertheless, the jury is still out. I’m hopeful, if not terribly optimistic.

GamesBeat: Raph Koster spoke at one of our events. His topic was all the lessons that we should already know before we start building the metaverse.

Morningstar: I have so much respect for Raph. He’s probably one of the best thinkers on this out there.

GamesBeat: Are there things you would pull from Habitat as lessons we should know now?

Morningstar: I wrote a paper called “The Lessons of Habitat,” so the answer is yes. A lot of the things that we started talking about after Habitat as core things we’d learned, I think people have largely internalized those. In particular the emphasis on the human element and the social element being where the hard and interesting challenges were. At the time there was a lot of focus on the technology, a lot of concern about 3D graphics or haptic interfaces. Now with metaverse you still have stereo optical headsets and things like that. I don’t think these things are the main event. Those are things which might possibly work someday, but the key challenges are in the social realm, in the realm of how people interact with each other, what they can do with each other, what they can do to each other, the ways their behavior impacts the people around them.

Even though this is something which has been–a lot has been written. A lot has been said. A lot of people have been far down these roads and brought back scouting reports. You mentioned Raph as a great example of that. But there’s still a tendency to fall into the trap of, “Wouldn’t it be cool?” There’s also a tendency for companies, particularly companies that are heavily invested in this, to focus on the product they want to sell rather than the product people want to buy. I have confidence that market forces will sort that out in the long run, but the question is how much churn and weirdness there’s going to be while that happens.

GamesBeat: It’s interesting how you can get a similar message from a very different person. I happened to have a speaker at our last conference who was a 20-something Roblox game developer. He worked with brands to bring them into Roblox. He said that the first thing that they say to him when they start working together and they have an idea for what to do–they always say, “Can you build us a town square for our brand?” And then his answer is, “Stop. We should build a game. This is Roblox. This is what you want to do.” It was interesting that they always had some impression that if you just built a gathering place, people would have fun there.

Morningstar: If you build the right kind of gathering place people will have fun, but it’s not clear that such a gathering place will have any connection to whatever your brand happens to be. Most brands are not in and of themselves a place that people will congregate around, at least outside some very special, narrow interest. A lot of the larger brands, in the sense of the dollar volume of the business behind them, those things tend to be more diffuse. They don’t tend to be things people identify themselves with. If you’re Coca-Cola, people don’t want to go to Coca-Cola Town. That’s just not a thing. If you’re doing something which is narrow and specialized and attracts a narrow community of people who are interested in that one thing, then that might well be a thing people would congregate around, but then it’s not going to be this big thing. It’ll have a brand manager behind it.

It’s funny when you talk about town squares. One thing that’s become a running gag with Randy Farmer, one of my collaborators on Habitat, when we’re looking at people’s demos of their VR systems or their metaverse systems or their virtual worlds–one question we always ask is, “Are there chairs?” Very often the mere presence of chairs is an indicator that the people doing this haven’t thought it through very deeply. Why does an avatar need to sit down? It’s more of a running joke than a serious critique, but nevertheless it’s proven to be a surprisingly effective diagnostic.

GamesBeat: How do you think some of the competition will go this time? There’s the forces of the walled gardens, and then there’s the forces of the open metaverse. Do you think the open metaverse can win?

Morningstar: I’ve always been very much on the open side of that. It’s interesting, because a lot of things that started as open systems ended up evolving into not exactly walled gardens or closed systems, but de facto dominated by one or a small number of key players, just because of market forces leading in that direction. One thing that’s interesting is that you can’t necessarily a priori predict who the winners will be. If you have an effort which is predicated on “Company X will be the winner,” if they don’t happen to stumble across the formula that makes it work, then you’re likely to fail.

Even if the end state is a walled garden, an open approach is almost necessary in order to discover what the viable commercial thing will end up being. Generally if you try to predict, you’ll get it wrong. The future is complicated, particularly with large, complex projects that take a long time to unfold. A lot of learning happens along the way. Freeman Dyson makes a great remark in one of his books. He explains that it’s not worth doing anything that has a time horizon of five years or longer on it, because by the time you get to the end of your five-year plan, the world will have changed enough that you’ll inevitably have done the wrong thing.

That doesn’t mean you don’t do something that’s more short-term and incremental aimed at a much longer-term objective. But if you start out with some kind of grand plan that’s going to cost gajillions of dollars and be your big push, those things have not had a good track record of paying off. It’s largely because of this learning element, plus the fact that the world is not static.

Particularly looking at what Facebook or Meta is doing, I think there is a strong element of this, “We’re building a thing we want to build,” as opposed to, “We’re building a thing people want to have.” They have a lot of money. They have a lot of resources. That’s not to discount the possibility that they may pull something off there. But I’m not sure this is a thing where resources are necessarily the determining factor. It’s a question of whether you can learn. It’s just historically harder for large, established entities to learn. That’s a fundamental lesson in Silicon Valley.

GamesBeat: People wonder if they’re painting themselves into a corner by betting so much on VR.

Morningstar: I’m very much a VR skeptic. Part of that is because I’m terribly, terribly prone to motion sickness. It biases me in that direction. But I’ve yet to find a set of goggles or anything that didn’t make me want to turn the thing off and go gasp in fresh air for the next half a day. I don’t doubt that situation will improve. Technology has definitely gotten better. But I don’t feel like it’s there yet.

The thing I find more exciting is the more transparent augmented reality concept. The most exciting thing like that that I know of right now is Tilt Five, Jeri Ellsworth’s project. I love Jeri and I love everything they’re doing. It actually works. You can put the thing on and it’s right there and it works. Another company that I think is interesting, although the jury is very much still out, is Mojo Vision. I’ve been following their founder, a friend of mine, for many years. That could be very cool. I don’t know if you ever read Vernor Vinge’s story Rainbow’s End. He has an interesting take on the future. His story True Names was a major influence on Habitat. Rainbow’s End portrays a future heavily entangled with augmented reality, which is very much the kind of thing that Mojo Vision might enable if it works.

Things which are much more permeable to the physical world, I think–my sense is those things are more likely to be successful than these fully immersive experiences. Once again, I’m not necessarily the world expert on any of this. And I know for a fact that I’m heavily biased, again, by getting terribly motion sick when full-motion graphics fill my field of view. Other people just eat that stuff up.

GamesBeat: I often find I’ve got opinions that probably have no basis in fact, that are emotionally driven. But it feels to me like the gaming people are the ones to figure this out.

Morningstar: That feels intuitively correct to me. Part of that is because game developers tend to be extraordinarily pragmatic. They tend to be very–let me say this carefully. Not ideologically driven. Not in the sense of political ideology, but in the sense of, they don’t start out thinking, “Everything must be X.” They’re more inclined to let themselves be pulled in whichever the direction the users are buying into. They’re not afraid to cheat left, right, and sideways when it comes to doing graphics. They don’t tend to be purists with respect to any of their rendering schemes or user interface techniques or any of that thing. They’re just down in the mud pragmatists. I think that will serve well in this situation where nobody quite knows what they’re doing.

GamesBeat: Jason Rubin had a similar comment. He said, “I have to believe that the metaverse will be built with a game engine. The people who know how to use game engines are game developers.”

Morningstar: The other thing is, game developers worry about what my former boss Steve Arnold at Lucasfilm Games referred to as the “funativity quotient.” Since they’re in the business of unabashedly providing an entertainment product, they’re not driven by an agenda to deliver any sort of experience except that which entertains people, that which keeps people coming back for more. This goes back to the pragmatism thing I mentioned a moment ago. Being really focused on what people like. What do people enjoy? What do people want to do? As opposed to, for example, what work they’re trying to get done.

A lot of business and commercial software ends up having really crappy user interfaces because the users don’t get a lot of choice or say in how it’s structured. They just have to use it because their job says they have to use this thing. There are lots of really obnoxious products you encounter in business which have that flavor. Whereas game companies don’t get the luxury of dictating to the users, “You must use this.”

GamesBeat: Have you had a chance to interact with many of these modern-day metaverse builders?

Morningstar: Not really. I’ve been disconnected from that world for a while. I go to GDC every year, when it’s held anyway, and meet up with the extended cloud of–you could call us the old fogeys at this point. I talk with those folks. But there’s a lot of stuff that people are doing that I’m probably blissfully unaware of. Not having been in the game industry for quite some time, or not having been first-hand in the game industry–I have a kind of foot in that camp. But the last game company I was involved with was 10 years ago.

GamesBeat: We touched on blockchain some. But I do wonder, if there is a lot of resistance from game players and game developers now, will we wind up bypassing blockchain on the way to the metaverse?

Morningstar: Once again, there are the blockchain true believers, who have this almost religious devotion where it’s the solution to all problems. Then there are the people like us who look at it as a tool that does certain kinds of things. If those are the kinds of things you need to do, this is a good tool to use. But there are lots of things it doesn’t do or it’s not the best tool for, and people are nevertheless advocating blockchain. The toxic one being voting systems. Almost nothing that blockchain brings to voting systems–you can talk to the voting security people, and they’ll say it doesn’t solve any of their actual problems. But blockchain true believers will tell you that blockchain is the solution to all your voting problems.

GamesBeat: We had a story today about DAOs and how it turns out that the control of DAOs ends up being highly concentrated.

Morningstar: It’s another artifact of the evolutionary path that a lot of these blockchain operations companies and projects have had, which is that they’re basically technology-driven, and to a lesser extent ideology-driven. They have perhaps been less firmly anchored in economic fundamentals than they ought to have been. One thing, to toot Agoric’s horn a bit more, is that we have a whole flock of actual real economists who are behind the design of our economy and our contracting framework. Their job is paying attention to these kinds of concerns.

People have this tendency, particularly people who have utopian yearnings, to say, “We’re going to build this thing and everybody will behave. Everybody will do X.” Those plans get disrupted when it turns out that everybody doesn’t do X. This was one of the very early lessons from Habitat. You can build experiences predicated on a model of user behavior, but if it’s wrong, those experiences will not work. That seems like a trite thing to say. Well, of course. But the reality was that early on, we designed things that didn’t take into account a realistic model of user behavior. When everybody didn’t do whatever thing it was predicated upon, it would fail. We learned the hard way to pay attention to that. That’s a lesson that people have to learn over and over again.

GamesBeat: Prototyping and playtesting.

Morningstar: Prototyping, playtesting, and don’t think that you as the developer are representative of your users’ behavior.

GamesBeat: If things work out the way Agoric would like them to, what do you see happening? What’s the great outcome?

Morningstar: The great outcome for us is a framework for–we talk about distributed contracting. That’s a little abstract. But basically enabling decentralized cooperation. Enabling people in far-flung parts of the world or in communities that wouldn’t otherwise interact with each other to be able to cooperate with each other in constructive and beneficial ways. A lot of opportunities for cooperative behavior end up–you get friction from trust issues, from infrastructure issues, from lack of technology. We’re hoping to fill that in.

The fundamental thesis is that if you lower the risk of cooperation, you get a more cooperative world. That’s our guiding ideal.

GamesBeat: If someone comes to you and says they’re going to build a modern version of Habitat, do you have any advice?

Morningstar: I’ve got nine generations of servers I built that you can have the code for! A piece of me says, “Why would you want to do that?” And there’s a piece that says, “Awesome, sign me up.” I know that I’m fascinated by the underlying technological and design and engineering and architectural issues around doing that. But I’ve also learned from experience to take that interest and apply it to building things that are actually useful for something.

Open virtual worlds, open world games in general, have been interesting, but I don’t see that anyone has stumbled across anything fundamentally new for a long time in that area. The question I would ask to someone who proposes something like that is, “What do you want to bring to the problem space that hasn’t been tried before? What do you think is a new wrinkle, a new solution?” And I’m not talking about better graphics or better display devices or more realistic physical models. I’m talking about something fundamental to the experience.

The landscape is littered with failed or just marginally successful systems that were undertaken with the best of intentions by very smart people, but ultimately didn’t go anywhere particularly important. I don’t necessarily believe that you have to know where you’re going. In part because we never know where we’re going. But you should at least have a vision for something that you’re bringing to the table that hasn’t been there before.

GamesBeat: I remember seeing some of the Habitat documents that were at the Oakland game museum.

Morningstar: Yes, the MADE. Randy and I have been involved with them for a number of years now. Under their aegis we’ve reconstructed Habitat. You can now play it on the web, the original Habitat, in a Commodore 64 emulator in your web browser, talking to a version of the Habitat server that has all the original server logic, although it’s new code based on a server platform that Randy and I developed a few years ago. But it’s the original client software, the original world experience.

The thing that astonishes me is just how good the Commodore 64 emulators are. Good and bad in the sense that they’re very faithful, but that means they faithfully replicate all the things about the Commodore that kind of sucked. But they do it very accurately. What you get is a very faithful reconstruction of what the original Habitat experience was, although of course you view it through the lens of 2022 rather than the lens of 1986. What you think is normal and particularly acceptable in terms of response times and framerates and all of that–it’s probably not going to be too satisfying in that respect. But there it is. You can play it. It’s cool. We have people who do that.

The MADE has collected a lot of materials. We’ve gotten releases for all the original source code, so that’s all available. We have a lot of the design documentation and other material. We’re working on getting release of the code for the successor systems that Fujitsu developed. We’re still in discussions with lawyers, so I don’t know how much I can say about that. But that’s something we hope to pull out. And some of the later things that Randy Farmer, Doug Crawford and I did at Electric Communities back in the ‘90s. Hopefully we’ll pry that out from the lawyer’s filing cabinet that it lives in. A lot of this material is becoming available. Anybody who wants to dig into it and ask questions is encouraged to do that.

GamesBeat: I was curious about whether Alex Handy was going to bring the MADE back.

Morningstar: They just reopened. I think it was just last weekend. I haven’t been up to visit them because I’ve been busy with family things, but I’ve seen pictures that Alex has posted. It looks like a really nice new location that they have. It looks like a better location for getting foot traffic into it. I’m cautiously optimistic.

GamesBeat: Is there a moment from the Habitat days that springs back to you as a fond memory, or something very memorable for you?

Morningstar: It’s hard to know where to begin. I have lots of stories from that era that have nothing to do with Habitat in particular. But I think some of the greatest moments, looking back, were when people started doing things that we never expected. In particular, some of the emergent social phenomena. The one that just absolutely floored me was people doing coordinated, choreographed song and dance productions using a lot of fussing around with stopwatches and talking to each other on the phone to get their relative latencies to the server calibrated, so that they could do a bunch of things on the screen with different people from different places all coordinated, so it looked like it was done in synchrony when in fact it was all wildly asynchronous and decoupled.

The fact that people would go to the kinds of lengths that were required, the kinds of obsessive attention to detail, to engage in that kind of creativity–it speaks to something very powerful about people’s drive to interact with each other and be creative. It took me by surprise, but in a very gratifying way. Even though the thing they were doing is ultimately, in an objective sense, kind of trivial, it was personally meaningful to them.

The key lesson I drew from that was to not get too fixated on what you think people should be doing. Pay attention to the things that are meaningful to them.