Collaborative Learning through Flexible Web CVE - Teaching through Multi-User Virtual Environments - page 479

each other in what we call an environmental
container. Then we had to design a XML-based
language, basically to describe the 3D scene.
This language uses the scene graph paradigm:
a hierarchical decomposition of the renderable
components in a scene.

On the basis of this consideration we described
the stage in the XML through a single model entry
as shown in Algorithm 1.
It is clear that a new world instance can be
generated simply by putting together and reusing
different parts already configured thus creating
new virtual sessions as composition of pre-existent
modules. WT04 Scene graph

Avatars The WT04 scene graph (Figure 4) is organized

as a sequence of 3D environments called parts
in which users can navigate moving from one
to another simply colliding with special interac-
tive objects, often in the form of ports or gates,
working as teleports, thus causing the unloading
of the current part rendered by the engine and the
loading the next one.
Since this environmental container can be
considered a mere sort of stage where the action
can take place, we can think at it as a whole,
monolithic composite geometry without any kind
of interactivity or behavior.

Avatars are graphical embodiments represent-
ing the participants in the collaborative virtual
experience. The WT04 schema describes avatars
using a particular node structure with the follow-
ing features:

1. Avatar Identification Properties

These properties describe how the particular
avatar looks like. We can set up, for example, the
avatar's skin, the avatar's vest (that is due to the
team membership) and the avatar's trousers. In

Figure 4. Top level scene graph XML Schema

Algorithm 1. Sample of WT04 scene graph, coded in its XML schema

<World>
+<Configuration>
+<Part num=”1” url=”@/Part/MeetingPoint.W3D” descr=”First stage”>
+<Part num=”2” url=”@/Part/TreasureHunt.W3D” descr=”Second stage”>
<World>