Liveblogging MIT: Virtual World Technology and System Design

Wrapping up the day was a panel on Virtual World Technology and System Design with panelists representing Linden Lab, The Electric Sheep, Croquet, and robotics and sensor studies from the MIT Media Laboratory. All hosted by IBM's Bob Sutor.

"Let’s remember that we’re at MIT. The Massachusetts Institute of Technology. That’s Technology with a capital T. The social aspects are all well and good, but let’s remember how things connect to technology. You may have connected with virtual worlds through various types of gameplay or onscreen, but let’s talk about when The Matrix gets here. Is anybody else excited about that?" opened Bob Sutor, setting the tone for the rest of the discussion.

Virtual Worlds Technology & Systems Design: A Panel Discussion

Moderator: Robert Sutor, Vice President, Open Source & Standards, IBM

Panelists:
-Cynthia Breazeal, Director of MIT Media Laboratory’s Robotic Life Group, MIT Media Laboratory
-Mark McCahill, E-Learning & Collaborative Systems Architect, Duke Univerity; Croquet Consortium Architect
-Joe Miller, Vice President, Linden Lab
-Jerry Paffendorf, Resident Futurist, Electric Sheep Company; Director, Acceleration Studies Foundation
-Joseph Paradiso, Associate Professor of Media Arts and Sciences, MIT Media Laboratory

Sutor: Let’s remember that we’re at MIT. The Massachusetts Institute of Technology. That’s Technology with a capital T. The social aspects are all well and good, but let’s remember how things connect to technology. You may have connected with virtual worlds through various types of gameplay or onscreen, but let’s talk about when The Matrix gets here. Is anybody else excited about that?

Sutor: Cynthia, you work in social robots, which I guess are better than non-social robots. Robotics and AI have long been a strength for MIT. Can you talk a little bit about the relevance to virtual worlds and how virtual world technology can improve robotics.

Breazeal: I was struck by the morning panels on identity with the high level of synergy. The twist is that what robotics gives you is a digital transformation back to the physical world with a different representation in a different location. Imagine that if you flew into the Eiffel Tower, you were also jacked into a tourism robot and could go around and interact with people in the physical world. A lot of questions boil down to master/slave issues. The robot will need a lot of local autonomy. It’s interesting to think how the robot’s intelligence will complement our own understanding and how to design robot avatars to convey what you want them to and not what you don’t want them to. These are the kinds of issues within the robotics world we’re starting to think about differently. Robotics isn’t just a tool to use at the distance, but a mode of communication.

Sutor: Mark, I understand that you coined the term surfing the Internet. If we extend this to the idea of multiple virtual worlds existing, will people surf them? Will they move between them? And how will the technology work? Will Croquet be the Apache of the future?

McCahill: It’s interesting because we have to talk about the idea of whether there are multiple technologies underneath or multiple visions for the same technology. I think we’ll also end up with multiple technologies for pragmatic reason, because the capabilities you may want will be available for technology A and not technology B. If you buy into the idea that there are different artistic visions, I think you have to buy into the idea that there will be different technologies. And that raises questions of how you surf them.
I don’t think that’s a huge problem, though. Maybe it’s a bad metaphor, but look at all the video players. Some players are better at handling video and others at voice, but they get bundled together with a tag asking which you want to use. I think you’ll see multiple worlds just for pragmatic reasons to find support for what you want, but that’s not a big problem for people using these worlds if there are clients that say “When I go to this world, use client A, and when I go to that one, use client B.” I think we should expect multiple technologies, like in video players or in the backend of websites. It also solves the problem of versioning. If some features are tied to versions, you either have to be backwards compatible forever, or you need to handle different clients.

Sutor: Second Life has undergone huge growth, but at the same time, Joe, you’re adding new futures. And one of them is voice, not just VoiP. What’s special with what you’re doing and how will it fundamentally change our experience?

Miller: Our first requirement is that we wanted to make it easy for your avatar to walk up to a group of people and have a conversation without requiring additional bandwidth or other technologies. An added requirement is proximal voice so that I hear people talking to me where they’re talking from. We also wanted to support this notion of communities of interest. Second Life supports thousands of groups of people who have chosen to join together for some reason, whatever it may be. We wanted to support voice so that you could engage in conversation with them wherever you may be. It’s a similar experience to party lines on the phone. The third idea is that we wanted to get private, high-quality conversations. As George Lucas said, “51 percent of a great movie is the sound track.”

Yesterday we launched all the aspects of that technology to begin to integrate it. What we’ve seen in the beta grid is that people are changing the way they interact. We’re plugging a massive hole in communication. If your purpose in Second Life is to collaborate, it makes sense to change it. We’ve seen richer interactions, richer roleplay, richer engagement between strangers all over the world. This rollout is the first step in a process over two months where we’ll roll it out over the entire grid.

Sutor: Jerry, are you changing your name to Electric Chipmunks?

Paffendorf: I’d love to see that idea eventually where you can change your brand every day.

Sutor: In looking at accelerating technological change, what do we need to do to optimize it? This technological change of virtual worlds, what more do we need, and what are the implications?

Paffendorf: I really have a creative background. I’m a technologist by thinking that technology is what drives change in the world. I’ve dedicated my last few years to hanging out with geeks and coders and learning where things are going. And that brought me to Acceleration Studies. Working with that group, I got to spend several years in close conversation with amazing minds, trying to organize the conference, and now taking what I learned there as a creative.  A lot of what I learned is a rational faith in accelerating change. There’s a proposition that we question whether the world is going to get as crazy as a lot of people say it is. What I fit in with Electric Sheep is saying that yes, we can rationally do things that we can’t imagine now.

Sutor: Joe, as we discussed beforehand, primarily my interactions with virtual worlds are with keyboards and mice. And that’s not that high tech. You have a project called Dual Reality. Can you tell us about it?

Paradiso: The idea is that eventually there will be portals to the virtual world all over the place. I’ve got screens of various sorts and goggles and floating balls that we’re working on.  A lot of my work has been in sensor networks, and a big question is that as you roll out these big networks, you see an explosion of data, how do you browse that? Getting that real world data into environments, and Second Life works well for us because we use buildings and architecture, but that’s a big challenge. It’s a kind of digital omniscience. Our sensors aren’t locked into the here and now, but they can bleed out into the world around us.

In terms of sensor devices as input devices for humans, keyboards don’t hack it. I’m intrigued by the way the Wii has succeeded. The game has taken off, but the industry was slow to adopt it for a long time. Eventually you’ll want displays and sensors all over the body, and a lot of it will be processing. The computer will fill in the holes.

Audience:  As with earpieces? Is that an example of a non-human-fluent product? You lose that ear. With the Wii you don’t lose anything. Where is that headed?

Sutor: Well you do lose the ablity to type.

Paradiso: The idea is that in dual reality you don’t lose, you add.

Audience: I’d like to ask you a question about when things might happen. I’m interested in being able to have meetings online where nuance like facial expressions. When will that be possible?

McCahill: It already is. As William Gibson said, “The future is already here. It’s just not evenly distributed.” HP has a product that allows for that.

Audience: But will people be able to use that? That seems like it’s only for special occasions.

McCahill: If this is really media, learning to play the media might be what we do. I remember in the early days when Macs came out and people had to learn to use mice. If the media is expressive enough, people will learn how to do that.

Miller: Part of the voice idea that we have is being able to create gestures based on your voice. Not at the bandwidth you’re talking about, but you can gesture a little more expressively when you’re making a point. I see a benefit in being able to take out soe of the telltale signs we have while communicating face to face.

Breazeal: You also have to think how the modalities complement each other. How the fact that you have expression in voice offsets the lack of some expression in movement. That’s a scientific question I think.

Paffendorf: It’s less sexy, but I’ve seen people in Croquet use separate boxes to represent different aspects of video or sound or movement.

Audience: Are there standards being developed for gestural assistance across systems?

Breazeal: A lot of people are rolling their own protocols.

Paradiso: In gestural studies, there is a vocabulary of sorts.

McCahill: That would be really crucial for carrying across worlds. If you develop an avatar that is similar to you and moves like you, you don’t want to have to relearn it every time.

Audience:  As this gets much broader adoption, as is happening, what are the standards for providing editing for international options so that you don’t offend someone unintentionally.

Paffendorf: I wonder how much of that will be technical solutions, like running algorithms to change the way I see you.

Miller: I think some of the cultural differences we have in real life get modulated away because there’s no reason to bring them to the table. I’m assuming that everyone in the world has the same tools, and they may use them in different ways, but we’re seeing a sort of leveling that way as in Second Life over half our users are from outside of North America.

Sutor: Just as a datapoint, where are most of your users outside of the US, Joe?

Miller: Believe it or not, it’s France and Germany, the two countries that are the most vocal about the experience.  We have no technical limitations of putting server farms elsewhere in the world to house the land. It’s easy for us to distribute those. Our challenge and what we’re dealing with now to get to Web scale is that the assets that make up Second Life from our residents take up 50 terabytes. We wouldn’t mirror that elsewhere and try to keep it current. We have an interesting next generation architecture that we’ll be talking about over the next 60 days and we’ll be expanding in orders or magnitude and concurrency.

Sutor: Do any of you see peer to peer playing a larger role? I’m not currently running a sim on my laptop, but in Snowcrash, that’s what he is doing.

McCahill: That’s what Croquet is, so yes. Another advantage is that you can work off the main grid and decide when you want to publish. It moves the land ownership issue down to the individual, so that allows for a lot more diversity in spaces and governance. So, yes, I think peer to peer is the way to go.

Sutor: And this will intersect the business model of who gets paid for doing what.

McCahill: And there’s a freedom of speech issue here as well. When you look at other peer to peer technologies, there wasn’t a single place to go and serve a warrant, for better or worse, btu that’s one advantage.

Sutor: Will DRM show up?

Miller: If digital rights management were a basic part of the internet, I think we’d have a whole different technology today. I don’t see an ubiquitous technology solution to DRM being valuable in virtual worlds today.

McCahill: But like it or not, the incumbent media companies are going to want to put it there.

Audience: It seems like the virtual world’s accessibility is limited to a small slice of the world population with a high quality PC and high bandwidth and time to play. Are there efforts to move it to venues like cell phones or other options?

Paffendorf: It also depends on what you think of as virtual worlds. If you wake up in the morning and all you have is a mobile phone to find out you’ve been paved over in a type of Google Earth type application, there’s an intersection.

Miller: We fully expect this to spread to a variety of other devices.

McCahill: The lesson of the $100 laptop is that it gets technology down to cell phone levels. It needs a better graphics processor, but you’re at the point where this can be free with a two-year commitment to the cell phone company. Someone just needs to do it.

Audience: Back to the DRM issue for a moment. In SL since you’re dealing with a global community of collaborative users, how do you get users in Germany to agree to a DRM agreement from the US down the line?

Miller: It’s a question we deal with. How do we organize a worldwide community of residents who live, work, and play together in everyday life, but they live in widely different area and under widely different laws. I don’t have good answers for you now. We know we have to work under the laws of the country organizing the platform, but at some point I think you’ll be able to contact a person in SL by calling their virtual phone from your land line,  so I think that in the long term these communities, on whatever platform, will be recognized as real, vibrant communities with their own governance.

Audience: Many people around the world percentage-wise have TVs. Do you see that as an access tool?

Paffendorf: That’s one of the things we work on, is figuring out how to get ready for interactive TV.

McCahill: Part of the thing is that once people get involved here, why would they want to watch TV? They’re the hero of their own stories in these platforms, why watch others?

Sutor: Of course we have much more computing power. The PS3 is a supercomputer in the living room. So as we look to move beyond what is a fancy home theatre, how will we continue to see our home augmented?

Breazeal: It’s interesting that from the world of robotics that everyday objects take on the quality of robots without us thinking of them as robots. Cars for example, or the Roomba. I think that trend is going to continue where robotics becomes an ubiquitous technology that is simply integrated. I think it will be not as much of a new technology, but an augmentation to our current technology.

Sutor: So if you think of augmented reality, there’s a lot going on in the physical world.

Paffendorf: I think of virtual worlds as a way of externalizing your imagination and the Web. It’s the Web everywhere world that looks kind of like Roger Rabbit.

Audience: As these worlds expand in power and size, what is the state of researching virtual, real ecosystems.

Paradiso: In some sense what we’re doing out in the lobby is similar. Physical sensors are fairly easy to build and then integrate to virtual worlds. Sensors become very generic. It’s what you do with them that’s open.

Sutor: Any final comments?

Paradiso: I’ve been really bothered these days by the Fermi paradox. The famous conjecture that when you look at the stars in the galaxy and galaxies in the universe and conditions for life, there should be life. But if that’s the case it should have been here. I think this is a crisis. It means either they’ve destroyed themselves or they’re doing something else. Maybe through virtual worlds, we see a glimmer of what they’re doing. Maybe virtual worlds will save the world.

Paffendorf: Going to inner space instead of outer space. You’re quantifying your imagination and making it tangible so we can all look at it. Then there are aspects of the virtual world where you recreate things. And I’m thinking of WoW and the 4D world. WoW is an excellent example of lifelogging where you’re keeping track of everything you do. We’re noticing a huge trned of that online of lifelogging. I think we’ll eventually cap out with billions of people online in a Google Earth situation, and we’ll be logging all of that.

Sutor: Okay, bouncing around.

McCahill: I think the UI needs to change. I can use it now, but not everyone can.That’s one thing we’re looking at.

Breazeal: I think of how much of that can be automated. Looking at aspects of robotics and AI learning from the user to integrate into it.

Miller: I’d echo what Mark said. I think it’s amazing the number of people who have been able to plow through a bad interface and stay. But there’s a huge barrier to entry right now. It’s much too complex. But we’re at the very early stage of this transformation. There are some amazing ways we can propel  this forward over the next years if we work closely together.

Joey Seiler
www.VirtualWorldsNews.com
joey (at) showinitiative.com
(512) 535-8650
skype: joey.seiler.vwnews

Comments

[joe public]

hmmm...and why was paffendorf on this panel again?

[Jerry Paffendorf]

Because IM IN UR WEB BUILDING UR METAVERSE.