David Anderson


For the past twelve years or so my research has been in the general areas of distributed and real-time systems, with a focus on developing middleware and toolkits to enable applications with person-to-person communication. Presently I am a research scientist at Mitsubishi Electric Research Labs (MERL) in Cambridge, Mass. During the two years I have been at MERL, I have been developing Spline, a platform for multi-user interactive environments, and Diamond Park, a social virtual reality application built on top of Spline.

Diamond Park is a social virtual environment in which multiple geographically separated users can speak to each other and participate in joint activities. The central theme of the park is cycling. Visitors to the park can explore a square mile of 3D terrain. In addition to the 3D animated avatars of human users, the park contains a number of computer simulations including a tour bus.

Diamond Park is implemented using a software platform called Spline (for Scalable PLatform for INteractive Environments) that provides comprehensive support for the key features of the environments we seek. Spline provides a convenient architecture for implementing multi-user interactive environments that is based on a shared world model. The world model is a distributed object-oriented database containing information about everything in a virtual world---where things are, what they look like, what sounds they are making, etc. Applications interact with each other by making changes in the world model and observing changes made by other applications.


There's more information about Diamond Park and Spline at MERL's home page at http://www.merl.com. My recent publications in this area include Anderson D.B., Barrus J.W., Howard J., Rich C., Shen C., & Waters R.C., "Building Multiuser Interactive Multimedia Environments at MERL," IEEE MultiMedia, 2(4):77--82, Winter 1995.

Barrus J.W., Waters R.C., & Anderson D.B. "Locales and Beacons: Efficient and Precise Support for Large Multi-User Virtual Environments," IEEE Virtual Reality Annual International Symposium, (VRAIS-96, Santa Clara CA, March 1996), IEEE Computer Society Press, Los Alamitos CA, 1996.

Anderson D.B., Barrus J.W., Brogan D.C., Casey M.A., McKeown S.G., Sterns I.B., Waters R.C., & Yerazunis W.S., "Diamond Park and Spline: A Social Virtual Reality System with 3D Animation, Spoken Interaction, and Runtime Modifiability," MERL TR 96-02, January 1996.


As for issues and topics for discussion, I am interested in trying to develop a list of standards that the group feels will be needed in order for large-scale virtual environments to be widely deployed, and to assess how far along we are as an industry in understanding what these standards should look like.

For example, Moving Worlds from SGI and others is about to be adopted as VRML 2.0. In terms of providing a shared space for multi-user interaction, VRML 2.0 provides for independent browsing of shared documents, but leaves open many questions, such as how the distributed activities in a shared virtual environment should be brought together -- i.e., what is the architectural basis for sharing that brings together the effects of having many users in the shared space, along with physical simulations, intelligent agents, etc? How can all the dynamic changes in the virtual environment be communicated in a scaleable way to the participants? How can very large virtual spaces be composed from smaller, independently designed and implemented worlds, and later fit together (hopefully in a seamless way) to form a new whole? The VRAIS paper cited above (available at http://www.merl.com/TR/TR95-16/Welcome.html) provides some ideas about how to begin answering these questions.

I'm also interested in hearing about the applications that workshop participants find most compelling, and in looking at what common needs they have. What is most hindering progress in getting these applications built and deployed -- is it technological limits such as high network latency, or lack of good tools for developing content, or underdeveloped standards, or something else?

David B. Anderson
Research Scientist
Mitsubishi Electric Research Labs
Cambridge, MA