After a morning discussing the ideas behind virtual worlds, augmented reality, and the future of work, the afternoon featured discussions from a technology perspective of the virtual world. Tony Parisi looked at what steps we need to take to work with Metaverse 2.0, Vladlen Koltun discussed the Stanford Virtual Worlds Group work with Dryad (a very neat collaborative modeling tool), and Jon Brouchoud looked at the tools we use in a social setting can be used to change architecture, the way we work together, and a health facility in Nepal.
Metaverse 2.0: “There I Said It”
Tony Parisi is working with Media Machines, but, unfortunately, there's nothing he can talk about there. Instead, he’s here to discuss the long view of the metaverse. As one of the co-creators of VRML (14 years ago last week), he’s been around the online 3D space as long as most people.
Now he sees us as “Metaverse 2.0.” Instead of talking about the visions of the metaverse, he’s focused on the business of technology.
“Is it soup yet?” he asked. While some, the MIT Technology Review, for example, look at the idea of a Second Earth as something to come, needing the integration of Second Life avatars with something like Google Earth. Will Wright, though, has this to say: “Metaverse? It’s already here.”
There’s 3D integration and Web connection today, but that’s not what most of us mean when we talk about the metaverse.
We’ve already gone through one boom and bust cycle of the metaverse. Project Habitat, Worlds, and VRML have launched and fallen down in the late ‘80s and early ’90s--primordial and stone soup. Now we’ve gotten into canned and talk soup.
Today we have hundreds of thousands in virtual worlds and one billion people using the Internet.
“I think there are two ways to break through the wall,” said Parisi. “We explore the long tail of the virtual world and put a little more Web into the virtual world.”
The long tail is a term popularized by Wired editor Chris Anderson to describe a phenomenon where the Internet allows non-hit books, movies, and more continue to sell long past their release, accumulating profit along the way.
His three rules for the long tail are to democratize access (but virtual worlds are too hard to use right now), democratize production (but virtual worlds don’t have many common standards to simplify production), and to provide filters by giving people ways to find content (and virtual worlds have bad search).
“We have XML. Build to that,” said Parisi. “We have the Web. Interface with that. There are basic principles of game design and user interface that we could take away. And there are lessons to learn from the last 15 years of Web development and interfaces.”
The Web took technologies like FTP and worked around it. Virtual worlds need to do the same.
“You might say, ‘But, Tony, 3D is harder,’” he said. “Yeah, you have to do real-time delivery, but did you have to reinvent networking and content delivery? No, you didn’t. I hope that takes a prominent role in discourse to come.”
A Guide for Design's Library of Babel
Vladlen Koltun founded the Stanford Virtual worlds Group, which recently made waves with its Dryad application, a sophisticated tree-modeling program that is, hopefully, paving the way for more collaborative tools.
Koltun opened up with a quotation from Jose Luis Borges’ “The Library of Babel.” The library is filled with all the important books in the world, but also everything else. Without a guide, the librarians despaired of finding anything useful. It would also outlast all of humanity.
“The Library of Babel is a design space,” said Koltun. “The concept of a design space was described by Daniel Dennet to describe a formal specification of a domain of possibilities.”
Each point in the high-dimensional is a description of one possibility. As the dimensions increase, there are more and more possibilities. But as the dimensions restrict possibilities, adding in more variables, the possibilities become more and more relevant to finding answers to the query.
“I want to show you how these concepts can be used for intuitive 3D modeling by casual users with no design experience, the kind of people you expect to participate in virtual worlds,” Koltun said.
Most models are done by taking virtual clay (or prims or empty spaces, etc.) and pulling it into the necessary shape. Dryad, though, is meant to be a guide for finding your way through the possibilities in the design space.
Users can find trees that exist and tweak individual parameters to describe the tree.
“But that’s a not entirely intuitive, and it’s a little bit geeky,” he said. “Not everyone adapts to using 98 sliders.”
Alternatively, users can scroll through a map of possible trees where trees closer to each other are more similar than those that are farther apart. As users zoom in, they can see trees representing small changes in the 98 variables.
Most trees looking nothing like trees, though. Valubale trees take up a small, small space. Dryad has a guide, though, based on a density estimation looking at a small set of landmarks to then assemble a function describing how useful samples are likely to be.
To find the initial set of landmarks, the group looked to the users of the tool.
“Whenever someone saves a tree in Dryad, the location of the tree they are implicitly voting for is uploaded to all the Dryads in the world,” he said. “Each tree becomes easier and easier to create.”
Since December, over 10,000 people have downloaded Dryad and placed over 2,000 landmarks in the trees, creating a reasonably accurate map. The modeling time for most trees is a matter of minutes. Many of the trees still looked nothing like traditional trees, though they are very pretty.
“We had no idea our design space contained such possibilities, but our users ventured further and further in,” said Koltun.
With the quick time, ongoing creative design, and collaborative approach, Koltun believes the work is applicable to casual users, virtual worlds, and additional applications. More information can be found at vw.Stanford.edu.
Next up is humans and buildings.
Building Together with Studio Wikitecture
Brouchoud is working on Studio Wikitecture, an attempt to use Second Life as a 3D wiki for collaboration in architecture.
Open-source copyrights, crowd-sourced information resources, and products based on voting are adding more and more creativity as they span the Web. Companies like Innocentive, which brings solutions seekers and providers together, have success because they bring together wide varieties of individuals in an open environment.
Wikitecture applies the same principles to build architecture in an open environment.
“If it works for products, perhaps it would work for cities and buildings as well,” said Brouchoud.
The built environment is too big for any single firm, architect, or organizing entity to grasp in its entirety. Since CAD models can be repurposed and the completion of a building can be broken down into steps architecture is a lot like an OS: complicated, remixable, and modular.
The Open Architecture Network provides a library of design and a platform for open-source atmosphere, encouraging others to modify many designs for their own needs, promoting evolution.
Studies have shown that the building industry is the most inefficient in the world, said Brouchoud, and it’s getting worse. Now is the time for collaboration that can promote efficiency.
But as wikitecture and the metaverse advances, it seems reasonable to see them come together.
Studio Wikitecture asked a Second Life architecture to design a meeting space, but without the ability to change each other’s work. It didn’t quite come together. When asked to build a courtyard allowing full modification along with a Flickr community showing off designs with comments and an archive to roll back, the courtyard came together.
“The technology is still very rudimentary here,” said Brouchoud, “but you can see that the build is much more coherent than the first experiment.”
Now the group is looking at a health facilities building for Nepal in a contest sponsored by the Open Architecture Network.
Working together the community assembled pages and pages of research. Then they started to build. The build has parcels for building and design, but, most importantly, it had a wiki tree filled with objects that users could use to design in the building parcel and resubmit the design back to the tree, creating a new leaf.
As a leaf becomes more popular according to the community vote (in SL or on the Web), they become greener and brighter. If a leaf gets to be dark red, the tree will eventually prune itself.
40 members worked on the project leading to 50 different designs and 200 votes. With three weeks to go, the group is wrapping up the submission.
“While we certainly hope this results in a successful entry in the competition, we’re trying to be mindful that this is just an experiment,” said Brouchoud. “I find it really exciting that something really does seem to be working here. And the technology is not just limited to architecture. It could be used by any industry or group based on 3D building.”
What if every citizen could weigh in on city planning this way?
“The sustainability problem is far more complex than any encylopedia project,” said Brouchoud. “And if we’re all part of the problem, we might all be part of the solution.”
When asked about how to improve the platform, without consideration of cost, Brouchoud said it wasn’t an issue of technology as much as community.
It was important to not just work with architects, but to bring in anyone that had a suggestion to offer, from building options in Nepal, suggestions in culture, and more.
The group is looking to release the wiki tree software soon, evaluating possible work on other virtual world platforms, and evaluating the ability to pull models from more traditional design software into the virtual world.
Comments