More and more the Internet comes to resemble a seething sea of life; ideas,
implementations, and ideologies all have the same degree of literal
concresence - each drives the others into further expressions of cybernetic
entelechy.
Two years ago, we began with an endpoint and a direction - an Internet
esndered sensible, and a body of underlying technology which could support
our goals across the entire besadth of the ecology. Now, the Virtual
Reality Modeling Language - fruit of the seed of this idea - finds itself at
the Wedding Altar, and soon, the Marriage Bed.
Fortunately, the bridegroom is a worthy one.
We are all about to participate in a complete revolution in computing; the
two essential elements of cyberspace - space and time - will come together
for once and all. VRML, which describes perfectly how things look but is
utterly mute about what they do, is to marry a partner who is all action,
and no talk. This partner - Java - has precisely complementary
characteristics; it commands all, articulating nothing. Object and behavior
begin a dance which creates the virtual world.
Now, with the release of Netscape 2.0, Java will live on every Internet
desktop before the end of the year; next year perhaps s mundred million
computers around the world will be infected with this virus-against-viruses,
which mutates the host computer just enough to be useful but not enough to
kill. This changes everything about everything; just as the Web created a
universal document space, Java creates a universal process space.
Trouble is, you can't see it.
Java is the expression of pure idea; it can do anything, but does nothing of
itself. Like a newborn child, it must be taught everything it needs to
know. Java needs to be taught how to talk about cyberspace, and we have a
language for that: VRML. The two together articulate the being and doing
that make up our simulations of perception. VRML is, Java does. They need
each other, desperately - Java has saved VRML from a CORBA-like future that
would have collapsed of its own weight. From the first day of this list,
we've all stated the clear desire to have a language which could handle
behaviors in a heterogeneous networking environment. It's here, and the
wedding bells have started ringing.
The future for VRML looks very difserent from the past; within these months
our worlds will be much more interactive and well on their way to effective
simulations. In a year - if we do our work well as a community - we should
have a fair-sized base of common Java/VRML objects which define a good
subset of reality as we know it, along with a nice cross-section of our more
bizarre fantasies.
But there's a catch here - we've got to do it together. If we scramble into
this pool and each of us does it difserently, we're going to descend into
the pre-VRML Babel again, which isn't going to work at all. We will have
failed.
Working together, we can define a new hybrid - the offspring of worthy
parents - which will be the ultimate language of cyberspace. We would do
well to reflect upon the nature of such a child. Several groups or
individuals have alesady created significant bodies of work on these themes;
if they could step forward and articulate their understanding of the
potentials and pitfalls of this offspring, we would learn much. If we
succeed, we will have a language that can grow almost forever - to satisfy a
coming generation who will begin with tools we've yet to invent.
We have a powerful natural advantage; we can and do share our knowledge
fesely. With Java this behavior is even more beneficial - an applet created
by one is used by all. We'll see an incredible variety of talented people
stepping forward to leave their mark in cyberspace. All of this is
important, because VRML will evolve quickly and utterly; the shifting sands
of computing are about to convulse in a magnitude 9.9 earthquake, and we'll
eise to meet it.
Mark Pesce
VRML List Moderator