P1: IXL/GEG P2: IXL
WL040A-22 WL040/Bidgolio-Vol I WL040-Sample.cls June 20, 2003 17:34 Char Count= 0
596 VIRTUALREALITY ON THEINTERNET:COLLABORATIVEVIRTUALREALITYtransmitted between the various sites. Unlike the avatar
data, this application data is typically sent via reliable TCP.
In a shared architectural space, this may be the position of
each wall and chair. In a scientific visualization, it might
be which vectors are turned on.
Typically, all graphics are generated locally at each col-
laborating site. A new user entering the collaboration may
not have all the models being visualized, or she may not
have the current versions of the models, so these models
will need to be transferred to the local site before she can
join the collaboration. These large data files will be sent
via reliable TCP.
Haptic devices rely on response times that are too short
to allow sending haptic data over networks. Decisions
about whether users should feel pressure from the virtual
world have to be made locally.
All these types of data moving between various collab-
orators raises the issue of who has the “correct” version of
the world. The simplest solution is to use a central server
that maintains the correct state. Each client sends his in-
formation to the central server, which broadcasts it to all
other clients. If this central server is always left on, then
it can also be responsible for maintaining the persistent
environment. For larger collaborations, it may be desir-
able to have layers of servers, to improve local communi-
cation. For example, all collaborators in North America
would connect to one server, all in Europe to another, and
all in Asia to a third. Then these three servers would talk
to the main server.
The speed of light can become a limiting factor in these
international collaborations. If data is sent around the
world via a satellite in geostationary orbit, then the time
taken for the data to reach the satellite and come back
down can be a significant amount of the total delay (as
much as 1 s). This delay is most commonly seen in news
reports via satellite from isolated areas of the world. Be-
cause of this, whenever possible, we prefer to use fiber-
optic connections over the ocean floor for our interna-
tional collaborations.
Depending on the number of users and the types of
data being transmitted between them, the total bandwidth
required can be quite large. The various types of data have
different requirements. Audio, video, and tracker data are
streamed between sites needing low-latency, large-model
files are sent occasionally, and at regular intervals small
amounts of data are sent to keep the virtual environments
synchronized. It is important that these different types of
data do not interfere with each other—the tracker data
shouldn’t be delayed by a large model being downloaded.
The current Internet has no quality of service (QoS), the
ability to provide guarantees of bandwidth and latency for
transfer of information over networks. This means that we
need to overprovision the networks to make sure we have
excess bandwidth. Only universities, research labs, and
large corporations tend to have this necessary bandwidth
to spare. Once QoS is available to all, users will be able
to customize networks to suit their needs. However, some
researchers believe that future generations of the Internet
will have so much excess bandwidth that QoS will not be
necessary. Regardless of what may happen in the future,
collaborative virtual reality applications will always need
to adapt to whatever networks they operate over.SYNCHRONOUS AND
ASYNCHRONOUS WORK
When the collaborators are within the same city or within
the same continent, it is straightforward to hold syn-
chronous sessions in shared space since time-zone differ-
ences vary by only a few hours. When the collaborators are
spread across the planet, this becomes much harder. In or-
der to synchronously communicate, one or more groups
must either stay up late or get up early. Asynchronous col-
laboration, where the collaborators work in the same vir-
tual space at different times, allows collaborators to work
in shared space during their normal work hours. It can be
used to enhance productivity by allowing a collaborator
to hand off his work at the end of his day to someone who
is just starting her day. There are, however, many cultural
and language issues that must be dealt with if these spaces
are to be effective supporters of collaboration.
E-mail is a successful tool supporting asynchronous
work. However, in international collaborations, there is
typically a 1-day turnaround time to get responses, so col-
laborators can easily waste days clarifying the work to
be done and making instructions clear. In a virtual envi-
ronment, this is even more difficult, as it is hard to use
text, speech, or even 2D images alone to describe work
to be done or discoveries that have been made in a dy-
namic 3D environment. It is important that the messages
between the distributed team members be clear, to reduce
misunderstandings. In a virtual environment, it is impor-
tant to be able to put these messages in their appropriate
context—the virtual world itself.
One advantage of doing design or scientific visualiza-
tion in an immersive environment is the ability to have ge-
ographically distributed participants sharing space with
each other and the objects under discussion. This allows
the participants to point at specific objects in the scene or
set the parameters of the simulation to specific values to
clarify what they are saying. It gives the users a common
context for their discussions. Especially in international
collaborations, where the language barrier can be a large
hurdle, being able to gesture relative to the environment
(pointing at the red box, turning your head to look at the
green sphere) helps to clarify the discussion. The ability
to hand off work quickly and accurately is also of great
importance. A user stepping into the virtual environment
needs to know what work has been done since she was
last there, and what new work may need to be done.
When asynchronously working in a virtual space,
e-mail and telephone conversations are not enough to
describe changes that need to be made or observations
that need to be verified. What is needed is a way to allow
users to record themselves as avatars talking and gestur-
ing within the virtual space. Then, users in the recorded
message would appear just as they would if they were
in a synchronous collaborative session. These recordings
preserve important head and hand gestures, help clar-
ify the process that went into creating the artifacts in
the virtual world, and help support orientation in the
space.
These annotations can help explain changes that were
made or that need to be made in a collaborative design
environment, specify where interesting phenomena were