Powered by
JSPWiki v2.8.2
g2gp 17-01-2009
View PDF

Previous | Next | Contents

Section 4: Collaborative Virtual Environments#

This section is partly based on the article 'VRML Multi-User Environments' by Rachael Edgar and Ben Salem (Edgar and Salem 1998), researchers at the Networked Virtual Reality Centres for Art & Design .

4.1 CVE Technology: how does it work? #

The World Wide Web operates on the client-server model. This means that, for any communication to take place, one machine will adopt the role of a server and the other that of a client. Servers are set up to serve formatted files (using extensions such as 'html', 'jpg', or 'gif'), and clients are set up to receive them.

3-D on the web is served up in the same way as other file formats. The file extension for VRML files for example is 'wrl' and, once the server has transferred a VRML file, the browser software on the client side is activated. The browser builds, or renders an internal model of the 3-D world and displays an image of it on the screen. Whenever the user changes viewpoint within the world, the image is re-rendered accordingly.

Collaborative Virtual Environments (CVEs) are networked virtual reality systems, enabling groups of people to come together in virtual space. CVEs actively seek to support human-human communication and collaboration in addition to human-machine interaction.The participants are virtually in the same shared virtual world and can interact with it and with each other, represented by their virtual embodiment, an avatar (see Section 4.2).

The concept of actually sharing a three-dimensional world poses a number of problems that other, static, types of web media do not. Actions, reactions, geometry, images, positions, orientations, and conversation all have to be shared and passed across the network. The simple act of tracking the position and state of a number of avatars and passing this information out to a number of users can cause immense problems.

CVE systems typically try to minimise the information that is exchanged between server and clients only to that which is absolutely necessary. When entering a CVE there is an initial download of the bulk of the world description and then short bursts of communication as information is passed to and from the server.

Designing worlds for collaborative virtual environments is similar to creating single-user 3-D spaces. Typically, the server requires some additional identification information in the 3-D description file. In VRML files, this could be as short as a single line specifying the location of the server system, typed in with a normal text editor. In principle, every VRML world could be made multi-user accessible. Therefore, the same guidelines that apply to virtual reality methods and techniques also apply to CVE content creation.

Previous | Next | Contents