Review by David Duberman
Earth to Avatars 1996 Conference
Daily Spectrum: Interactive Media & Online Developer News
29 October 1996
Reported, written and edited by David Duberman
for editorial/ subscription inquiries, email The Daily Spectrum
Editor’s note: We interrupt regular Daily Spectrum programming to bring you
a report on a truly portentous event that just took place in San Francisco.
Report from Earth to Avatars Conference
by David Duberman
Something really momentous happened last weekend. Not many people knew it
was happening, but those that attended the Earth to Avatars (E2A) conference
felt it deeply. Subtitled Contact, Culture and Community in Digital Space,
the conference could also have been called the Online Community conference,
because that was the basic theme. Of course, a lot of other stuff was
covered, including artificial life, 3D graphics, theater and gaming, but it
all centered around connecting and living with other people in cyberspace:
Can it be done? If so, how? 2D or 3D? And much, much more.
Organized by Bruce Damer of the Contact Consortium,
with yeomanlike assistance from Lynn Macias and a host of others, the
conference was a tremendously exciting and inspiring convocation of bright,
imaginative and forward-thinking individuals and enterprises involved in
shaping the future of how we will interact with each other in cyberspace.
Below you’ll find a collection of descriptions of and sound bites from the
various sessions I attended, followed by brief descriptions of some of the
products on exhibit at the conference. I apologize in advance for any
misquotations and disconnectedness; I was writing as fast as I could, but
the speakers were even faster.
In the opening keynote, VRML co-founders Mark Pesce and Tony Parisi traded places at the podium every few minutes to provide their takes on recent history and the present.
Parisi: This show is more about people than process (in comparison to most
Internet shows, which dwell on the technicalities).
Pesce: Habitat was the forerunner of today’s online virtual worlds. It was a
distributed virtual environment that ran at 300bps on a Commodore 64. Snow
Crash (Neal Stephenson) gave me real vision and direction; WebEarth is my
version of that vision. The avatar is a vehicle of communication.
Parisi: VRML was the inevitable outcome of the need for standards. VRML 1
implemented 3D visualization for the Web. It was simple, feasible, and was
based on something that already worked. However, it was hated by twitch
gamers, interactive developers and multi-user people. "VRML sucks!" was the
mantra of 1995.
Pesce: The only thing that fits through the keyhole into the ether is your
self. Your avatar is shaped by the force of your actions, and is preserved
in the memories of the avatars you’ve communicated with. Mark went on to
cite a quotation from the book Cyberspace: First Steps by Michael Benedict,
which referred to actions as being both persistent and cumulative.
Parisi: The VRML Architecture Group (VAG) was a benevolent oligarchy, and
the great thing that came out of it was that the SGI spec (out of which came
Moving Worlds) got really good. The VRML 2 spec adds animation, interaction
and multimedia. Its foundation is an open architecture, so it integrates
with Java and other standards.
Pesce then talked about the work of Michel Foucault, whose book The Birth of
the Clinic records the process of how a patient, in the course of his
journey through a clinic, is supplanted by his dossier. The relevance to
today’s situation is that as designers contruct avatar protocols, they build
us all into a conceptual prison that must inevitably trap us. He described
about Foucault’s design of the panopticon, which was an octagonal prison
with a central guard post from which vantage point one could see into all
the cells. Foucault also envisioned that this type of structure could be
used for workhouses and schools. Mark sees cyberspace as the ultimate
panoptic environment, in which any point can see the whole, and the whole
can see/be any point. As we construct our selves in cyberspace, we can
exchange ego identities as easily as avatar bodies.
Obviously, Mark’s been doing a lot of deep thinking on this topic, and my
sketchy notes don’t begin to do justice to how well he expressed himself at
the conference, but hopefully this will inspire you to read some of his
writings and attend his talks.
Applications for Virtual Environments
The next panel, What in the World Do You Do with a World?, covered
applications of virtual environments. Panelist Zara Houshmand of Worlds Inc.
talked about how we’re so overwhelmed with text today, and that text is not
the way to communicate in cyberspace. She also cited MSN’s Comic Chat, a 2D
online world, as an example of a failed concept, because its randomization
of body language undermines the chat experience. She stated that the
multi-user experience in 3D space with role playing enhances educational
possibilities, and cited the Starbright project, which links hospital wards
containing seriously ill children via high-bandwidth links. This project was
mentioned several times in various sessions as a shining example of the
benevolent possibilities of cyberspace applications. Starbright helps these
children overcome isolation at a critical stage in their social development,
letting them communicate with each other, not just doctors and parents. It
also helps by giving kids a choice: If they’re relatively healthy-looking,
they can use videoconferencing, but if they’ve been ravaged by chemotherapy
or other treatments, they can use avatars.
She went on to talk about Eastern culture, in which face-to-face encounters
are more highly valued than in the West, and how 3D virtual environments
could facilitate such meetings, citing as an example the use of such
technology in Japanese dating clubs, where couples meet as avatars first.
Another interesting application would be business meetings, in which
workers’ avatars could be dressed casually as perceived by each other, but
in business clothing as perceived by a client. Houshmand quoted a Worlds
Inc. marketing slogan which states that "Reality is the best metaphor." This
was disputed by a number of other speakers throughout the conference. The
Palace’s Jim Bumgardner offered as an example of taking advantage of the
possibilities of cyberspace the ability to turn into a carrot to make a
point. The consensus opinion, though, seemed to be that people are most
comfortable in familiar surroundings (e.g., sitting in chairs). Another
interesting point Houshmand made was that VR is currently closer to theater
than to film or interactive multimedia; in fact, the latter remains
painfully close to its roots in computer-based training.
Also in the "What in the World …" panel, Hollywood director Brett Leonard
offered thoughts on his particular area of interest, which is how to address
the mass audience. This is a crucial issue that is not considered by many in
the industry, but I’m not sure if Leonard’s going about it the right way. He
cited as proof of his qualification to opine on this topic the fact that
over a billion people worldwide have seen his film "Lawnmower Man." For
better or worse, this was the first introduction to VR, albeit a passive
one, for millions of people. He claims to be working as a viral infiltrator,
introducing this technology into mass culture via passive media. Leonard
made a couple of interesting observations. First, he repeatedly used the
phrase "connect the dots," in connection with the idea that to draw people
in, you must exceed a certain production level. He also said, "Chat is
chaotic. It doesn’t empower me to do much. It’s like going into the middle
of a busy street and screaming." This struck a chord with some attendees,
and was cited on other panels. Finally, Leonard stated the importance of
giving people the tools for directing their own movies, claiming that
"Everybody wants to be a director."
Living Worlds
Another panel included participants in a consortium that’s developing Living
Worlds, an technical initiative to define a conceptual framework and specify
a set of interfaces to support the creation and evolution of multi-user
applications in VRML 2.0. The LW effort aims at a single, narrow,
well-defined goal: to extend VRML 2.0 to support applications that are both
interpersonal and interoperable. For example, if you visit Art, Betty and
Chuck in their virtual living room, and bring a virtual Monopoly board, and
you’re all using different systems, how do you know if they’ll be able to
see the board, game pieces and money, much less be able to play and follow
the rules?
If you’re interested in learning more about this, visit the Living Worlds Proposal. For now, I include some excerpts from the Living Worlds documentation as of October 25, 1996.
In its startup phase, work on Living Worlds has concentrated on three
questions:
- What new conceptual components (if any) will a distributed VRML
- architecture require?
- Between which components are standard interfaces needed?
- How can we achieve an optimal mix of standardization (to ensure
- interoperability) and openness (to leave room for innovation)?
Work on Living Worlds is governed by these principles: build on VRML 2.0;
standards, not designs; architectural agnosticism (the central server is not
necessarily god); respect the role of the market; and require running code.
The requirements fundamental to any interpersonal/interoperable VRML
environment include the ability to:
- insert objects (e.g., avatars) into a VRML scene at run time
- track and communicate changes to objects in real time
- allow objects to be driven by users in real time
- let imported objects be persistent
- protect the local scene from damage by imported objects (e.g., Betty’s puppy shouldn’t be able to eat the furniture)
- support persistent roles (for people) and rules (for scenes)
- exchange information among objects
- link objects dynamically to external (non-VRML) data/functions
- specifically, to support authentication certificates (e.g, dice, business cards)
Among the advantages of supporting interoperability among features lets
users take avatars and objects to different worlds and lets developers focus
their efforts on building single components (e.g., avatars, portable
interactive games) as well as combine best-of-breed components to yield
richer results with less effort.
Many big names are involved in the Living Worlds initiative, including Black
Sun, IBM, Fujitsu, SGI, Sony, as well as just about every supplier of 3D
models, and if you have any interest in setting standards and/or getting in
on the next big thing, you’d do well to check it out.
Interface Design
Several interesting questions and ideas came up at a panel on Designing
Interfaces for Virtual Worlds, including: Which is better for communicating
in groups; 2D or 3D spaces? Many 3D spaces are antisocial, in that they
follow the Doom model, which is essentially an office-type environment
redesigned for killing. 3D spaces are more compelling. AlphaWorld is better
than Worlds Inc. because you have the ability to stake out your own
territory and build your own space. Textual communication can be more
freeing than voice, which in turn can be more freeing than face-to-face. In
inventing a new language for communicating in cyberspace, we need to let
users define their own gestures rather than selecting from a limited preset
menu. A general theme during this panel was: "How much control over the
interface design do you give the user?," and the consensus was that that
users must be allowed to fail in order to learn how to build the best
interface. One important task of cyberspace builders is giving people tools
to create interaction. It’s also important to provide more versatile
interface devices than the two-dimensional mouse and joystick. One possible
solution was provided by exhibitor ITU Research of Cupertino, Calif., which
showed a prototype of the TouchCube. The cube-shaped controller consists of
five 2D position and force sensitive panels, which provides 3D position and
orientation input commands, plus a sixth sensor on the base for conventional
2D (mouse-type) input. Contact ITU at 408-446-4597; fax 408-446-1493; email
taizoy@aol.com.
Virtual Worlds as Theater
One of the liveliest panels was Virtual Worlds as Theater, hosted by Mark
"Spoonman" Petrakis. After a theatrical opening invocation of sound and
movement by a self-professed "wild woman" (sorry, didn’t get her name),
Petrakis spoke about how good avatars can enhance the ability to
communicate. Also, the intent of true characters is to guide us to
self-discovery. Analogies were made between the theatrical possibilities of
cyberspace and movies, which are giant puzzles in which each piece has been
created by a different person, and the novel, which presents a landscape of
the imagination that’s missing from virtual worlds, as in the reader’s
thinking, "What happens next? What if …?" Also, comic books provide a sense
of time in the transition between panels that virtual worlds designers could
learn from, and they define their rudimentary realities in quick strokes.
Well-known designer Brenda Laurel spoke about how we’ve been robbed of
ritual, spirituality and humanity, and how using avatars may allow us to get
them back. Finally, we can, believe it or not, learn from TV shows and
advertising such powerful capabilities as selling social attitude and
providing cultural resonance.
There were many other terrific panels, including Worlds for Gaming and
Fantasy Role Playing; Kids in Digital Space; Virtual Worlds in Marketing and
Electronic Commerce; and Digital Communities and Culture. In fact, many of
the best conflicted with each other, which leads us to believe that
cyberspace would be a great venue for a future such conference. Of course,
there’ll never be a substitute for real-world interfacing.
Sculley Closing Keynote
One "famous" participant in E2A was John Sculley, who is scorned in the
industry as a former sugar-water salesman who ran Apple for a few years, but
never bothered to learn much about computers. Thus it was perhaps surprising
to some that he gave one of the conference’s best talks, at the closing
keynote. Admittedly, there’s a need for the blue-sky dreaming that comprised
much of the conference proceedings. But for 45 minutes, Sculley held the
audience spellbound as he professed his lack of technical orientation, but
nevertheless provided a brilliant summation of the conference’s events and
concepts in practical, down-to-earth terms. He started out by describing the
conference as a "seminal moment," and our current status as "one of those
important discontinuous changing times," in which the defining metaphor
shifts from the empowerment of the individual (i.e., the personal computer)
to looking at the environment created when people network together. He
recalled how, shortly before he died, Jonas Salk spoke of the metabiological
revolution, which leverages the socializing experience of connecting minds.
Sculley further defined the nature of the discontinuity by talking about the
ongoing reinvention of marketing, which has transformed from selling to the
art of owning the customer relationship (e.g., branding etc.). He described
the Internet as the information supermarket, not superhighway, and gave us
the new three C’s: community, commerce and customization. Are you tired of
hearing about intranets yet? Get ready for Sculley’s extranet, which is made
up of different companies’ intranets, connected for secure
business-to-business transactions. What’s left after this transformation
(e.g., plain old HTML Web pages) looks like CB radio, while the current
banner-type advertising looks like junk mail. The Web is in great need of
reinvention, and that’s where members of organizations like Contact
Consortium can offer greater value. Giant media companies are moving onto
the Web with great vigor, not trying to make money directly from a Web
presence, but leveraging it into their total marketing pictures. And the
phone companies have given up the idea of being content providers, having
figured out that the ISP business is where they belong. In the new Web
model, the big money goes to those who offer big bandwidth as well as
quality experiences, and the latter is where cyberspace developers, many of
whom are moving over from the CD-ROM world, can profit. The online
socialization experience might integrate intelligent agents with humans, and
chat communities can become increasing important as forces that drive and
shape economies.
Sculley finished his talk by describing the four companies he’s currently
involved with. First, and to my mind, foremost, there’s RealSpace, founded
by QuickTime VR creator Eric Chen. RealSpace has created a new extension to
VRML 2.0 for adding photographic environments and objects to virtual worlds.
In a nutshell, it combines QuickTime VR-type technology with VRML. The
extension, called Image Worlds, allows panoramic (and spherical) images,
video and image based objects (IMobs) to be used in the creation of virtual
scenes and avatars. Currently RealSpace offers its spec and the free RealVR
IRC browser at its Web site. The company also
sells a Mac or PC CD-ROM for $299 with the RealVR Xtra for Director 5.0 with
panoramic viewing and VRML 2.0 real-time rendering, plus utilities to
convert QTVR to RealVR, as well as a wonderful tool called Vistagrapher that
stitches multiple images into a panorama. Sculley has also bought into his
son’s Seismic Entertainment of San Francisco, which creates virtual
exploration/adventure experiences (e.g., a trip under the polar icecap or
through Mayan ruins), also described as "neat content." Sculley’s Live
Picture demo’ed its FlashPix technology, in which zooming in on an image
shows greater detail instead of blocky pixels, and LiveWorld’s mission is to
create and operate high-quality original programming with a focus on
community, audience participation, time and events. The latter’s first
service, called Talk City, is now online at .
Also present at the conference in one form or another were the following:
Universal Worlds is an open Java/VRML 2.0 API developed jointly by
Misubishi, Chaco Communications, Velocity Games and Worlds, Inc. to provide
a complete distributed multiuser communications infrastructure for complex
virtual worlds. Find out more at .
Microsoft showed two virtual worlds, both available on its MSN commercial
online service. The 2D Comic Chat is notable for cute graphics, but more so
for the clever Emotion Wheel interface element, in which you determine your
character’s expression by moving the pointer towards the edge of the wheel
to choose increasingly intense degrees of coyness, happiness, laughing,
shouting, angry, sad, scared and bored gestures. And V-Chat lets people
communicate online from within a 2D or 3D multimedia environment using
bitmapped avatars. Users can make their own avatars and express themselves
with preset gestures. Microsoft offers developers the V-Chat building kit,
which enables the creation of a new 2D or 3D space from a template, or via
any program that can output VRML.
Much smaller than Microsoft, but promising some impressive technology, was
Princeton, New Jersey-based KATrix, which is developing what it calls
NeurRule Intelligent Agent technology. An outgrowth of research in robotic
control theory and artificial intelligence, it attempts to copy key features
of human motor control and skill acquisition. The core components consist of
Limb Coordination, based on inverse kinematics; neural networks for
human-like learning characteristics; rule-based control based on a
goal-directed inference engine and knowledge base; and agent class
libraries, which provide an object-oriented framework for building
intelligent agents. Contact KATrix at 609-921-7544; fax 609-921-7547; email
lane@katrix.com.
Sven Technologies showed off its Windows-based SurfaceSuite of 3D editing
tools by digitizing attendee faces and quickly applying them to a 3D model’s
head without distortion, thanks to the company’s original algorithm. Sven is
also working on a low-cost 3D digitizer, a tool for splicing surfaces, and
more. Find out more by phoning 415-852-9242; fax 415-813-1775; email
hello@sven-tech.com.
OZ Interactive demonstrated technology that
enables "agents with attitude" which inhabit 3D communities, interacting
with and offering help to human visitors. The agents incorporate mood
parameters, giving them distinct personalities which let them reply to users
based not only on the content of what they say, but also on moods evoked by
key words. The company also showed OZ Virtual, multi-user VRML software that
lets participants interact socially in any VRML world on the Internet as
their own Persona (OZ avatar).
3D Web (415-956-9730; mike@3dweb.com) showed Spinner, a $99 VRML 2.0 tool
for adding animation (i.e., spinning motion) to any 3D scene. Scheduled for
available in Q4 is Spinner Pro, described as a professional-level authoring
tool for 3D animation development.
CONTACT: cultures of the imagination is billed as an interdisciplinary
conference that brings together social and space scientists, science fiction
writers and artists to exchange ideas, stimulate new perspectives and
encourage serious, creative speculation about humanity’s future, on-world
and off-world. It takes place March 7-9, 1997 at the Sunnyvale (Calif.)
Hilton.
Integrated Data Systems of Savannah, Georgia (hotline@ids-net.com) offers V-Realm Builder, a Windows-based VRML
authoring tool with drag-and-drop functionality, multiple editors,
object/texture/material libraries, integrated video links and more. The
company also has a server and a free multi-user VRML browser.
Intel Distributed MOO 2.0 is a Java-based software framework for building
extensible virtual environments.
Coming in March, 1997 is a book by conference organizer Bruce Damer
entitled Avatars! Exploring and Building Virtual Worlds on the Internet.
It’s a step-by-step guide to the most popular virtual worlds including The
Palace, WorldsAway, OZ Virtual, Utiopia and others. Visit publisher Peachpit
Press or call 800-283-9444.
This January in San Francisco, SGI and Seybold Seminars will present World
Movers: The VRML 2.0 Developers’ Conference. Phone 800-488-2883 or
415-578-6900; fax 415-525-0199.
Last but not least, the full E2A conference proceedings are available on
videotape and audiotape. Write to the Contact Consortium at 343 Soquel Ave, Suite 70, Santa Cruz CA 95062-2305. And next year,
plan on attending this not-to-be-missed event, which will probably overtake
SIGGRAPH as the industry’s coolest conference.
