05 July 2007

X3D browsers, viewers, players, modellers, exporters, etc.

Every since I stumbled upon VRML and X3D I wondered about a middle ground of features between a VR professional development environment and a VR browser. It is one thing to be able to read and render a scene graph or play animations but it is another thing for an application be able to author nodes and script animations for others to view and play.

The 'end game' for my grant work has me specifying an application and network that let's a user interact with a scenegraph (to actually reposition and reorient objects to start with) but record and save some or all of those interactions to a new scenegraph or overwrite the existing one. This will be memory intensive if you consider a 80-year lifespan of reconfigurations and reorientations. Remember, X3D is supposed to extensible and so survive generations of computer hardware and operating systems. This feature is not recording the trajectory of the viewer or rendered scenes in his/her virtual world. Although, may be this feature would be of value to some, it's equivalent to the history feature in html browsers. This feature is time stamping and recording the avatar's interaction with the nodes, like turning and translating objects in his/her's VW. And gravity is turned on so things get crushed, roll, fall off objects and dropped to the earth (sink through water if present) if nothing strong enough brings it to rest. Electromagnetic changes of objects but not biological and nuclear interactions would be simulated as a function an avatar's input too. These sort of features are Computer-Aided Engineering type of features that will challenge the handheld device's CPU but not a server's if the VR is access via a handheld computer. Distributed computing design options will be addressed near the end of the grant after a static analysis of computation and network can be performed.

Serving up innovate displays of patterns of behavior and just trends about changing attributes of interest over user-defined objects will be new features to make VR a powerful tool to a decision maker. For example, a homemaker or corporate IT manager might set a goal (w/ a date and exact quantity) for a behavior of him-/herself or others that affect his/her VR. The goal could be the repletion or depletion of a class of objects (i.e. a maximum rate regardless of absolute quantity). The application would serve up a transparent 2-D graphic overlaid a zoomed out view of their entire world once the criteria for notification was met. Maybe a click to acknowledge the event would be required to clear the graphic.

1 comment:

Anonymous said...

People should read this.