VRML–Viewing Cyberspace in Three Dementia
By Mark J. Jones
CyberStage Communications
Toronto, Canada

VRML, the Virtual Reality Modeling Language created by Mark Pesce and
Tony Parisi, is a technology whose proponents vastly overestimate its
potential, while those outside the VRML community vastly underestimate
it. I sit somewhere comfortably in the middle. I have worked in VRML as
an editor and producer, and the conceptual and discursive possibilities
that come with creating 3D interactive spaces on the Web are truly
enticing and challenging. Of all Web technologies, VRML presents perhaps
the greatest chance for creating new innovative spaces for mass
distribution, because the flat interface dominating the ubiquitous Web
suddenly turns into a space of three dimensions. Yet, for the most part,
people just aren't using VRML even though browsers and development tools
have been available free for some time. The overriding question in the
minds of many artists and VRML developers has become, "Why haven't
people caught on?" Even the industry is beginning to back off; SGI
pulled out of developing VRML tools for anything but their own platform
as part of cutbacks announced in October of last year.

Why would a technology with so much potential be so widely ignored by
much of the Web community?

CyberStage's own VRML piece is a case in point. In February of 1997,
CyberStage launched Construct(s) and Meme-ing, a VRML essay authored by
Stephen Matsuba and hosted by the CyberStage Live website
(http://www.cyberstage.org). In the piece, core arguments are presented
not as traditional text, but as a series of virtual spaces and visual
images. You, as the "reader," explore the author's argument by moving
through different virtual environments. The basic structure of the essay
is comprised of five components: an introduction, three main points, and
a conclusion. Each component is presented as a separate VRML world, and
is downloaded as you move through the essay. The premise is that once a
person has "played" in each space, he/she would pick up on points in the
author's argument, and would in theory be able to support or refute the
overall assertion of the essay, much as if it was written in traditional
text.

In terms of editorial development, approach, mode of communication, and
the presentation of argument, Construct(s) was a truly exciting and
innovative project. The piece's aesthetic was determined by Matsuba's
belief that in new media, the current milieu of work is created out of
the linear-text book model.

Yet, since its launch in February 1997, the piece has had about 800 hits
to date, a relatively low number in spite of a large marketing campaign
(for an art piece), a two-week gallery installation with an opening, a
mini lecture tour, and listings on key VRML link sites. Even being
listed on the "SGI VRML of the Week" site added only 50 or so more
visitors. This is typical for a VRML site, and many people are growing
increasingly disappointed by the size of the these audiences. VRML will
certainly not disappear anytime soon, but the lack of reaction from
those outside its community does have some interesting implications.

One of the other key problems with the discourse of the VRML community
is that it assumes that the rest of world is where they are, or at
least, not far behind: the rhetoric is sometimes evangelical. This
premise encourages sweeping generalizations that leave others wondering
how often they themselves dabble in a little RL. Witness, as one
example, a statement by Carl Francis DiSalvo, in his ON VRML essay: "By
the (obviously inherent [sic]) virtual nature of VRML, our construction
of space is limited only by our imagination. The VRML environment is
only like architecture or theater as long as it functions as if it were
actual, rather than virtual" [RHIZOME CONTENTBASE, 11.7.97].

Anyone who has ever tried to create anything artistic on a computer
knows that we are bound by much more than our imaginations, and that
overcoming those challenges is an inherent part of the creative process;
tell a VRML designer that he is only bound by his imagination, and he
will laugh as he tells you how each browser reads his code in different
and particular ways. Likewise, the notion that theatre hasn't been
experimenting with the integration of so-called "virtual spaces" shows a
lack of understanding of what experimental theatre has been about since
the early 1960s; theatre has been experimenting with temporal and
cognitive spaces long before any of us donned HMDs (head mounted
devices).

Likewise, VRML co-creator Mark Pesce himself has made several statements
about VRML's role in the unification of the human race that are
grandiose and unrealistic: "It's necessary for the Web to grow well
beyond even its current boundaries; I insist that it is the initial
instance of a new form of communication, probably as important as the
birth of human language a million years ago."

Perhaps still in the comfort of 2D HTML, our text-based culture isn't
quite ready for VR on demand, at least not until a viable "killer app"
appears on the market that shows both its usefulness and its uniqueness
to a wide variety of people. Computer games are, of course, the
exception, and many successful games such as Doom and Riven use VR-like
sensibilities (first-person point of view, simulation of 3D space). But
even they aren't marketed as virtual reality, perhaps in recognition of
VR's relative obscurity to the general public. VRML may be the first
elitist platform on the Net, not necessarily because of access issues,
but because of consumer choice.

At the moment, discussion is critical for understanding VRML's
appropriate uses and values–VRML is, after all, relatively new. One
thing is clear: VRML will not attract the same audience as did HTML, at
least not yet. Therefore, VRML must prove its relevancy to the greater
Web community; otherwise, despite ongoing updates to the VRML spec and
the harping of those who have a personal and financial stake in its
success, it will go the way of the 8-track.