Video and audio support is handled by the Lucent Collaborative Video (LCV) system ( Civanlar and Gaglianello 1998), an IP-based, multipoint, video conferencing solution which supports both unicast and multicast protocols. The system is a hardware and software platform for personal computers and supports high quality video (MPEG-1 and Motion JPEG) and stereo audio from both live NTSC/PAL sources and stored multimedia files. An LCV encoder is able to encode two different resolutions of video, 320×240 (CIF), and 160×120 (QCIF), and transmit from 1 to 30 frames per second. An LCV hardware decoder allows a recipient to receive and display up to 20 different video streams, all at 30 fps. Each participant can choose the frame rate and size of the received videos. A dedicated audio bridge handles audio and mixes the requested input streams into a single output audio stream for each receiver.
In a virtual jazz club, for example, small groups of friends may gather around a table for conversation while listening to the music of a band multicast over the Internet/Intranet. As in a real-world club, members entering the virtual club can identify groups of friends already present and move easily from one conversation to the next while still enjoying the entertainment
An important aspect of real-world clubs is their spatial layout. Members, in most cases, are seated at tables while socializing and listening to a musical- performance. The act of joining a table of friends is similar to joining a group. Therefore the jazz-club becomes a set of multiple sub-groups; each group focusing on the interaction with itself, along with the general focus of some central event. The club table can be seen as a type of spatial boundary -- a place that contains its participants within a social group. As ( Mynatt et al, 1997) observes, "Spatial boundaries support and define social interactions and the development of social networks".
Just by the nature of sitting at a table in close proximity to others, means that a member can consciously communicate in a way that only those at his or her table can see or hear, i.e., either by talking quietly, side-comments, gesturing, or facial expressions. A member's discreet gestures may only be noticed by the members at the table, and may not be noticed by groups at adjacent tables. This sense of partial privacy (partial because the table is still in open view, a sort of open-boundary) makes social interactions more comfortable for the members -- this also applies in on-line situations because spatial-boundaries conform to the users' real-world expectations.