Supernova: Virtual Life or Virtual Hype?

This was curious: some fantastic panellists, but the conversation was long and looping and a bit directionless. This often happens of course when you have a lot of questions from the audience, rather than a moderated interview; nevertheless there are some good soundbites in there, I think.

As ever: not a transcript. Nope. There are huge chunks missing - Supernova is recording these, so if you have 90 spare minutes in the future, you can watch the whole thing. These are my longform notes, and I apologise to the panellists if I've misrepresented them anywhere, it's entirely possible.

The audio situation was bad: no mikes on the panel, and I was at the back sandwiched between Big Coughing Guy and Man Rustling Crisps Packet.

Virtual Life or Virtual Hype?

IBM lady, hosting:
Clay Shirky, essayist
Raph Koster – Areae
Rueben Steiger - Millions of Us

IBM: Is it life or hype? Is it something that will have real business benefit? How will it evolve? The future comes from the fringe, and as it goes mainstream, it goes through a metamorphosis: what form will this take? But first, intros...

Clay: I teach at NYU, and I blog.

Rueben: I’m the CEO of Millions of Us, an advertising agency specializing in virtual worlds. 

Raph: I’m president of Areae.

IBM: (Opening gambit)

Raph: Near as I can tell there are 35 million people using virtual worlds and almost all of them are playing games. There is a rapidly growing but proportionately small group using them for things other than games – advertising, user creativity, business.

There’s a whole lot of hype surrounding 3D internet, and 3D is a red herring. There’s a lot of hype around metaversey things. A lot of it will come true .. The real estate biz is going to be huge in virtual worlds. But right now we’re in a bubble.

There’s misunderstanding over what their real power is.

Rueben: we’re seeing thematic indicators, but the pitched battle over whether this is REAL or HYPE is... I’m trying to see it as a non-issue.

A lot of the debate focuses on weird things like Second Life's numbers... but the fact of the matter is, we think in 3D. It’s inconceivable that 3D worlds will replace the internet, but people like these environments, they like them because they’re fun, 90% of what they do is socialise and have fun, and primarily it’s about community and social connections. 

It’s an adjunct to established online communities.

Clay: The phrase virtual world creates a category... we’re looking at immersive social 3D spaces; the big […] is games. No non-games have achieved large scale adoption, broken only by Second Life numbers. 

The thing driving adoption of virtual worlds right now is not the realism of the virtual world, but the visually-appealing-enough thing plus enough... (analogy with AI here)... specific use cases. 

3D environments are not headed to general purpose, but to very particular uses of these environments. Games are the first and most successful of these. 

IBM: Real Estate? 

Raph: Virtual world is a perfectly fine category because all it means is a virtual sim of space. 

We’ve had virtual worlds in text since the 70s. So 2D, 3D, text, whatever, the client is irrelevant. 

Given you have a virtual world sim space, the question is, what do you put in it? There has to be stuff to do. Without that people will go away, with it they’ll stay. Some of the to-do is just looking at the space. Hard to beat 3D for visualizing 3D spaces! 

I want to tour the house that I’m thinking of moving cross-country to buy. Picture Google Earth where you can hang out and check out the neighbourhood? That’s a good app. It’s not a repeat visit type thing, the play pattern we see in games, though.

Chat is not useful, btw. Chat is too available. Chat is not enough. 3D space often has limitations on chat, like how many people you can engage at once… so this is a hammer and nail problem. Here’s a hammer! Let’s pound on heads, let’s crack eggs … but we have to focus on SPACE and what we can do with it. 

IBM: Stickiness?

Rueben: I can talk a little bit about the environment that I understand, which is Second Life: the interesting thing is that the mantra is, it’s not a game.

But it IS. Life is a game.

The set of use cases or goals are just extraordinarily broad. The game is the experimentation... making money, that’s the game. Etc. 40,000 people are logged in having fun with that game. With respect to stickiness, it’s a hard game to get into. 

Most folk don’t get it. They don’t think it’s fun. We notice that – and linden has yet to solve – there’s a really high attrition rate until 3.5 hours of use, then it drops wildly. Well, what happens at 3.5 hours of use? 

Raph: You meet someone! 

Rueben: right. You meet someone. So... funnel users to content? Abject failure. Host a dinner party? Aha!

A lot of reductionists will just say, this is a chat room with an interface. It’s a lot more than that, but yes the critical factor is when you meet someone. The social element. 

If I dropped you out of the sky in Prague, you’d hate it. You don’t know anyone, you don’t speak Czech.

It would suck. That’s what Second Life is like.

Clay: the attrition is in excess of 90%, but it’s falling, the success at getting people to log in twice in a month and that’s falling. This is even more worrying.

Linden leaves behind almost everyone who tries it, and I never bet against the users.

I’ve never bet that the users don’t know what they’re doing. 

This could be the time that I’m wrong, but…I think what we’re seeing is sorting into Special Cases. Games are special cases. I think games are cognitively special. Gamelike situations cause different responses to non-gamelike situations... what I think we’re seeing in the general purpose virtual worlds is a discovery of which purposes other than games will be effective. Real Estate does seem useful short term.

Rueben: So your assertion is, don’t bet against the users. But users are saying, based on stats coming out of Linden Labs, is that they [inaudible].

I’d like to press... I would say that the total registered users is a red herring. The important stat, if you want to bet on users, is total hours of system use and total …

[inaudible]…average use per user is decreasing. 

Raph: all of those metrics are crap. Registered users is like, a crap metric. Money spent and hours per users are measures of devotion, not measure of adoption in any form.

The only decent metrics are, of recurring users, what % is coming back month after month. We call that monthly uniques. How many didn’t make it through the download? The threshold for download for anyone on the net is 5mg or less, or they go away. How many stay for the first five mins? How many convert – stay and then come back?

For a free to download product on the net, having a conversion of in the 8% range is PHENOMENAL. That’s triple the going rate. In the free-world it’s good; in the pay-world, it’s crap, a tenth of what it needs to be.

But there are few industry standards out there, we have to extrapolate. The Linden graphs look wrong to anyone working across virtual worlds, due to the way they’re structured.

I would not use money. 1% of users can be monetized can be 10x to 100x the baseline user. So the amount of money is a measure of passion not growth. 

Rueben: it’s frustrating... in things like mmogchart. Everyone is compared based on number of subscriptions. If you take total hours of system use, 16-18m hours a month, and compare that to WEBSITES, it would be in the top ten. 

Look at the MMO numbers. 60,000 subs... Tiny... what you have going on is a different paradigm, you have a bulk of people as free users. They contribute to the system, and then the 8-9% of people who are premiums get an average monthly value of 30-50 bucks. [inaudible] 

Raph: the answer is somewhere in between. Average play time per week in game world is 20 hours a head. 

Rueben: yeah we used to discuss these silly things like, when will it tip. We’ll tip at 100k. This is all moot! We tip when my wife logs in, because she’s NORMAL. By those criteria we tipped 6 months ago.

Q: we’ve just talked Second Life numbers for 25 mins. You can do better. I beg you to get out of the Second Life bucket.

IBM: It’s early stages. This is generational. Base it on a history of gaming; I’m going to come back to ecosystems, but let’s talk about tipping points.

Rueben: one way of evaluating this is to say... our company, we’ve been around 1 year, we work exclusively in Second Life and we’ve just chosen to expand past Second Life. We’re going into Gaia Online: it’s web based, it’s a virtual world, the average user spends an hour a day, and has the 2^nd biggest message boards on the net. 

[Generally] we’re moving from 2D to 3D. Social to entertainment. 3D entertainment, we have WoW. 3D social, Second Life. 2D social, Gaia Online. 2D entertainment, Mafia.

You could argue that the virtual world universe arrays itself in this quad.

As a company, Second Life monetizes not well at all. You strip away the metaphor, it’s a hosting model. In a forward looking sense, shouldn’t the operating company of a virtual world be in the biz of selling objects? Second Life isn’t in that, and they’d make a TON of money if they were. 

Clay: virtual reality. The locus of collaboration. We’ve been having that conversation since the videophone in 1964 at the worlds fair. MeetUp, a most popular category was people in Ultima or Everquest. 

To the ecosystem question: I don’t think this is early days. I think we’ve seen enough of these visual environments. Communications are not a perfect substitute for transportation. We’ll look to hybrids. 3D representation of the actual 3D world... [inaudible] 

Raph: We have to look at history here. Leaving aside videophone and playdoh, virtual worlds of 1978-9 were games. Continued as games. Commercially available as games in 1982. Invented in Britain btw. Became more social and went graphical in 1985.

That was the first ("Habitat") virtual world, and it didn’t go very far. By 1989 we had a handful of virtual worlds that had 1/10^th of users of games, and that pattern hasn’t changed. The game industry... went worldwide. 

Having this question in Korea would be moot. The entire population is in the game world, and no-one is in the social world.

Q: Are you excluding Cyworld?

Raph: Cyworld is MySpace with a colorform apartment. One guy can’t chat to another in an apartment.

Rueben – but it has space, avatars and objects. 

Raph: I'd argue it’s not an avatar if it can’t talk. 

But... there’s a fundamental philosophical question here: are these things going to swallow the web, or be widgets on the web? Is one going to swallow the other?

Snowcrash thinks it’ll swallow the web, and it’s wrong. These will be widgets on the web. 

IBM: this is on top of the internet. Where will the keystones be? What will they be?

Clay: what do you think needs to be different from today, is what you’re asking? 

IBM: yes

All: it’s not going to die.

Rueben: the ultimate insult is ‘it’s a widget’

Laughter 

Rueben: it’s probable that in terms of big issues, the biggest [?] is interoperability. What’s a probable outcome here is that Second Life continues to grow and perhaps becomes the predominant user-created world. An enormous ecosystem of alternatives – culturally and geographically relevant – and more polished alternatives will spring up, offering a more gamelike, directed experience. 

All of these things’ relationship to the web is that there’s no way they’ll supplant the web. 

We’ll never sort large datasets in 3D. The web ain’t going away. For things that are synchronous and social, you’ll tunnel into these environments. At that point, it begs interoperability. 

While users have a desire to have federated IDs, and go from world to world, platform operators have the opposite. The people who will lobby for interoperability will be the less successful ones. 

Raph: Users don’t want federated ID. People go out of their way to maintain multiple IDs. 

Rueben: No the concept is you’d like to be able to get utility from cash investments in one to [inaudible]

Raph: that’s a really narrow [...]. People like having multiple IDs.

Q: Hi, I’m identity woman. We’re seeing open ID take off in 2D web. People DO want to cart their ID around to some extent. They don’t want 100 logins. They want a manageable number under 10. 

Clay: That suggests that federated ID is defensive: it’s not about exploring ID; it’s about not having to have tons of logins. So there’s utility – logins – and there’s avatars/identity, which is much more expressive.

I think people want shared login, but do they want shared avatars? Walking your avatar out of starwars and into warcraft? That’s like going to a costume party in the wrong gear.

Q: going back to the application of real estate. You’ve talked about the substitute for transportation. And the use for collaborative work. But there’s another aspect, where Second Life has leapt ahead, and that’s in the architecture for real estate. I can design and build a model house in Second Life or another VW and allow people to tour it. 

Clay: The answer to Rueben’s interoperability question is going to come from tools like SketchUp. Users create an account in a world then create something in that world. But   with SketchUp... people create, then later can bring their creations into environments.

This is like Lego. Creating something in SketchUp and wanting to put it in multiple different environments... 

Rueben: the reason is hasn’t happened so far is that when Second Life was developed, they used prims instead of mesh. So the existing universe can’t be imported effectively. Things can’t be imported in either, it’s like in The Fly... things turn up as inverted baboons. 

Collaboration is extremely powerful in these environments. Not better than real life but powerful.

Five of the biggest tech companies have platforms under way for enterprise-use cases of virtual worlds. You’ll have realtime translationso that linguistically different teams can communicate.

The bar is not to be better than reality, but to be better than web apps. 

IBM: yeah, our team is worldwide, and we meet regularly in Second Life.

Q: we’ve been talking like the tech we’re using now is THE tech, but it’s getting better all the time. I’m seeing innovative stuff here that’s really [life changing][long description]; when you move into the world of the next generation of this stuff, and the next generation after that, don’t you think we eventually get to somewhere where the 3D stuff will be so good that some of the things you say might in fact change? I think we’ll get to a place where social worlds become way more important because the experience is so much more interesting. 

Raph: Yes... but what you’re mentioning is the unrelated field of VR. There is no overlap. It’s an accident of history that virtual worlds and virtual reality share so many similar […], they overlap not at all. 

VW practitioners have no idea what defines telepresence, or who Jonathan Stewart is. The virtual world stuff came down on what telepresence is way better than virtual reality ever did.

Virtual worlds are communication, not representation. It’s not about representation! 

The representation will get better and better and better, but... the poster child for this right now is that the largest shared spaces on the net have the worst graphics. I don’t think that might change.

Social spaces are about the bandwidth of communication, not the bandwidth of pixels.

Literally the only divider is, does it simulate space? If facebook had North South East West on its links, I’d count it. 

Rueben: there are companies who do what we do. “Metaverse services companies”. Not a single client has a clue what it means. We call ourselves an ad agency. We see virtual worlds as a form of social media. It’s emotion, social connection, what matters, the representation is to an extent arbitrary and tangential. 

Facebook? It’s not a virtual world. But they’re selling used virtual objects: gifts. You can give sushi to people. That’s a very social-worldy thing. 

Clay: if I stand on a chair, and I wanna get to the moon, I’m not there. If I get a ladder, I’m three times as close. But it’s the wrong way to get to the moon. Screens are the wrong way to convince people they’re in a world. The vision of virtual reality can’t be done with screens. We need neural implants for that. So there’s no poly count that‘ll get us to the point of believing we’re in that space. 

Raph: Except people can believe if they’re in a room when they read a paragraph of text. 

Clay: Yes, if you can get the brain to work towards belief, you don’t need polys. But to work straight up, you'd need implants. 

Q: The world is 3D, but we have 2D eyes. We don’t see in 3D; it’s trickery. Presenting people with 3D environments is hard: great for a game, but... 

Rueben: you remember these early systems where people would take the web and build these 3D environments... “would you like to visit 3D Yahoo!”? Memory is encoded in 3D... 3D has some very powerful characteristics. 

I don’t spend an overwhelming amount of time in Second Life, but - I will never remember an IM conversation, but if we were having this meet in Second Life, the way we’d remember it would be very interesting. We’d remember physiognomy, and position. 

Raph: And giant rabbits. 

A lot of virtual worlds' mutation recently has been size and scale. Historically that’s been a bad choice. Even though they’re about space, they’re better simulating segmented space. Zones. 

You don’t want to hear people talking if they’re not in the same zone as you. No-one necessarily wants the worlds to get bigger. 

Rueben: people desperately want emotional and social things. Find and communicate.

Raph: …and something fun to do. The best virtual world operators are those who realize that their job is not so different from running a real life venue: a bowling lane. A bar. Some place with something to do. 

Who goes to a bowling venue only to bowl? Those who go to a bar just to drink? Those are sad cases. Both cases are to go with friends. The place is a social lubricant. Virtual worlds are exactly the same: Slaying a dragon, or building a house: it’s an excuse to hang out. If you get there and there’s nothing to do, your friends will want to go bowling instead. 

IBM: next generation [inaudible] 

Raph: Anyone who is not a gamer today is an aberration. They’re weird. I work with this generation. Every online human under the age of 15 is a gamer and completely comfortable in virtual space.

Because of that, they’re savvy. They will sample 20 or 30 virtual worlds in a row. They are not loyal to worlds, they’re loyal to their social circles. Their social circles are formalized: named, even trademarked!

They move in social units in Dunbar-sphere sizes. This is why socialisation by itself is never the key – the social bonds migrate. This generation is loyal to each other and the network of friends. 

Clay: That process happens in parallel in communications: the idea of their being AN environment containing all the tools, well, that’s AOL. They’re in more than one virtual world at a time. They’re joining forums, smsing, voiping.

They’re making collage constructs of software. 

Some acceptance of that you’ll never have 100% of the users’ attention – and accept that other applications will be [included].. [needs to happen more]. 

Rueben: As systems change, the concept of stickiness has changed. In one kind of virtual world, you can invite your friends. Import your Gmail. People can up and leave immediately.

In other systems, there’s a diabolically sticky aspect in that people are bound by the fact that their friends are there. Their content is there... it’s not transportable! And [inaudible] is there. That combination of network effects is … it’s hard to migrate people. 

[…] 

Raph: There is a huge amount of nostalgia for their first world, their first love. The first world they try. It’s an interesting phenomenon.

Rueben: In Second Life, the early users were very much virtual world fans. They were there to explore different identities. Increasingly people are there and they create themselves, they’re what I call augmentists. This is increasingly exciting. It’s going to be interesting to see whether the 3D equivalent or friendster-or-fakester takes off... 

Raph: The longer they stay, the more they’ll engage in identity play, not less. It’s really common to replicate yourself, and then later people are like, wait a minute. I could be X..! 

Richard Bartle argues virtual worlds are exactly FOR this. He says the people leave when they’re figured out who they are. 

Rueben: you don’t think it changes when this becomes the 3D equivalent of email?

Raph: No. You’ve all done this. You’ve gone to a bar one night and worn something weird. Right now I’m in authoritative lecture mode. That’s really different to how I am in other circumstances.

Q: What are the correlations or differences in a business perspective, because so far this has all been about people? 

Raph: ask IBM. 

IBM: it’s grey. 

Raph: my sense is that in the end, people prefer to play. Even when they start doing business, engaging in really serious things, people want to play. For whatever reason. 

Partly because of the affordance in the space. That guy cited architecture: I think it’s collaborative product design leading into the fabbing industry. That’s REALLY exciting. 

Rueben: typically we’ll do a project for a company. Some interesting things always occur. Most of our projects are in the marketing realm. You’ll get some results in Second Life and much bigger results on the web. 

But the idea of the virtual world becomes really infectious. People love to think about it! The first part of our process is brand translation: take Diageo. The drinks people. We want to be in Second Life! They said. We said, why? Your product is meaningless there. There’s no actual drinking. 

So we created a bar where when you take a drink, something unpredictable happens to you in the form of an animation.

Audience member: all the avatars suddenly look much better to you?

Laughter.

...Heh. Exactly. And the drinks themselves have a social network: if I toast with my drink, everyone else’s drink does the same. 

In doing this translation process- the end result is cool – but it's the introspection going on in the translating that’s super interesting and unanticipated. When you ask people what they’re really really about, they’re often not really sure. It’s fascinating. 

Q: I come from Korea, and virtual culture is more advanced, I think. [Long story] I feel like this culture is slowly coming over to the west…

IBM: leapfrogging. Do you see any [inaudible] for leapfrogging?

Raph: Leapfrogs often happen under particular circumstances. Korea leapfrogged in large part because of some rather unusual circumstances: the Asian financial collapse followed by a government-wide decision to wire the nation. Dense population in high-rises. A cultural tradition of respecting games: games are for kids here, but there, a Go master is like, wow. The enmity with Japan, the import restriction on game consoles to Korea... all of this is a crazywild set of circumstances conspired to make Korea this [amazing place for virtual worlds.]

Clay: and meanwhile America is behind in terms of broadband because our copper wire network was so advanced that it took longer to replace...

[…]

Raph: There’s a confluence between photosynth and googlemaps and zillow and mapquest. These guys are rushing into a major collision, and virtual worlds will be part of that. Fabbing is esoteric, but interesting.

I think the confluence on games is going to get bigger: as far into the gaming culture as we think, we’re nowhere near the limits. NO WHERE NEAR. 

Games will become the dominant entertainment medium for this century. 

We don’t get what this means yet. In Korea the number one bestselling pop music star wants to hang out with the number one pro gamer - so that the gamer’s fame will rub off on the popstar.

This sounds like science fiction, right? It's not.

Within 5 years I predict it will be as normal to have your own personal virtual world as it is today to have your own blog. It’ll be odd if it can’t connect on a mobile device. We know it’s going that way, we can see it. 

Rueben: super interesting in terms of trends: everything is a game. Product design and experience design: the world is catching on. 

[Long inaudible section] 

Raph: my turn to restate: a huge side effect is the gameification of the real world. A lot of web 2.0 wouldn’t exist if those coders hadn’t grown up with Pong. Rankings, ladders, rep systems: these are straight rips from games. 

Clay: as long as we’re on predictions, I’ll predict that the commonest use in biz will not be collaboration, it’s going to be education. Mirror neurons. Accurate enough to train. Games do train reflexes beautifully. 

Rueben: prediction? Omigosh. Erm... my prediction is that the impact of virtual worlds and virtual object exchange, because in many ways, you mentioned Cyworld, what virtual objects is doing is encoding social hierarchies... this is a prediction that often sounds apocryphal, but people are going to replace a lot of the buying patterns that currently serve a human need (shirts, shoes, houses) – they’ll replace a subset of that by buying $2 shoes for their avatar. 

Q: in 5 years, what do you think the state of the art of the tools will be? Will 2D be blended with 3D? 

Clay: to put the skeptical argument out, I’m observing that the most successful uses are not just gamelike, but actual games. I’m expecting places where the real world can be gameified to be the most successful. So things like training. 

5 yrs? I think lots of things will be incrementally better, I can’t see a big breakthrough coming, although if it does [inaudible]. The social expectations of younger workers will be big part of the chain. 

Raph: games are really good at training, but better at teaching problem-solving. Collaborative problem solving in a shared space. I think there are lots of potential apps for this in a business context. 

The reason games work is because they’re inconsequential – it’s a loaded term, and I mean that literally, it has been proven a zillion times that you can teach effectively in a game, but you can’t grade it, because as soon as the kids realize you’re using it educationally... [inaudible]. 

Businesses are going to have to adapt. Their world is going to have to fit to the tools. Because the stats on the tools are mind-blowing. The educational potential of games is ASTRONOMICAL. 

Rueben: I draw a distinction between games and virtual worlds here... lessons learned from 30 projects... some of the most interesting projects have been the most humbling: usually to do with the synchronous, social, collaborative nature of virtual worlds rather than the playing environment of games. 

We’ve created great, polished stuff that doesn’t get a lot of use. The market will begin to construct their projects with this sort of stuff in mind: projects and initiatives are going to allow users to have input because users want to do that. Users don’t want to come and witness, they want to interact and change it.

On the part of the operator, it’s a very basic community principle. You train people to know what to expect of you. 

Q: how much international... 

Raph: of the largest virtual worlds are not in the west. The largest in the west is Habbo, which is European. The interesting thing is each of these was born […]... one of the most distressing things of virtual worlds is virtual hate crime and cultural rivalry. Language gaps are huge parts of this. 

Second Life is acquiring Euros much faster than Americans, and that’s leading to interesting tensions. The net encourages tribalism. Korean vs. American players in counterstrike or starcraft. While you can somewhat hide your race, it’s harder to hide gender. Impossible to hide your language. 

[…] 

Rueben: this just reflects reality?

[Out of time]

Not ending with a neat wrapup there, what with the gong going, but some really nice, interesting ramble. Thanks to the panellists!