1994 VR Conference Proceedings

Go to previous article 
Go to next article 
Return to the 1994 VR Table of Contents 

Teaching, Technology and the Human Spirit: Exploring Access to the Virtual Campus

By: Carl Brown, Director
High Tech Centers Training Unit of the California Community Colleges
21050 McClellan Road
Cupertino, California 95014
(408) 996-4636
E-mail: cbrown@netcom.com


For students with disabilities, particularly significant orthopedic disabilities, much of the difficulty associated with successful completion of post-secondary education involves the daily effort of getting to the campus, making the sometimes lengthy trek from one classroom to the next and coordinating the services of aides and other support systems to meet personal needs while on campus. What if students with or without disabilities could access a virtual representation of their college campus? What if arriving on campus was as simple as logging into a computer? What if walking around on campus was as easy as typing a simple keystroke or issuing a voice command? What if students and faculty could meet in a virtual classroom, in real time, from anywhere on the planet, without ever having to leave home? Each of these hopeful visions are a daily reality on DeanzaMUSE, a full-scale, text based virtual representation of De Anza Community College in Cupertino, California.

Motivated by an on-going interest in emerging educational technologies, a strong commitment to the future of distance learning and a willingness to explore new avenues for student access and success, De Anza has begun a bold new experiment in the future of virtual reality, teaching and learning. This paper examines some initial efforts at providing access to the DeanzaMUSE project through the use of "high-end" speech recognition technology.


In recent years, Internet, a global computer network originally conceived as a tool for scientists and educators, has grown and developed in ways perhaps unimagined by its original developers. Internet traces its beginnings to the early 1980s as a restructuring of the U.S. Department of Defense funded ARPANET computer network and a project to connect several hundred university and U.S. government mainframe computers (hosts) for the exchange of electronic mail (e-mail), information, and computing resources. Since 1986, the number of computer hosts on the Internet has grown by approximately l00% each year. By January l993, Internet was linking more than 1,300,000 hosts in most major countries. The exact number of people who access Internet computer services is unknown; estimates range from 8 million to l5 million people worldwide excluding users on hosts that, for security reasons, are invisible on the Internet system.

As a part of the deregulation philosophy of the Reagan administration, guidelines restricting the use of Internet were modified to allow commercial organizations as well as private citizens to use the network. Effectively, anyone with a computer, modem and access to an Internet account was free to use the resources of the world's largest computer network. And what resources there were: university libraries, enormous databases, access to government documents and research papers, global electronic mail and real-time conversations with other Internet users from around the world.

In the early 1980's Richard Bartle and Roy Trubshaw of the University of Essex began development of a new and novel form of Internet usage. Based loosely on the popular adventure game Dungeons and Dragons, they produced a real-time, text-based, multiplayer environment called a MUD or Multi User Dungeon.

Dictionaries on Internet terminology define a MUD as follows:

MUD: |muhd| [acronym, Multi-User Dungeon; alt. Multi-User

Dimension1. n. A class of {virtual reality} experiments accessible via the Internet. These are real-time chat forums with structure; they have multiple 'locations' like an adventure game, and may include combat, traps, puzzles, magic, a simple economic system, and the capability for characters to build more structure onto the database that represents the existing world.

European students using academic networks quickly improved on the original MUD concept and produced several new MUDs (VAXMUD, AberMUD, LPMUD). This second wave of MUDs (TinyMUD and variants) tended to focus on social interaction, puzzles, and cooperative world-building rather than combat and competition.

In 1991, the first MUSE (Multi-User Simulated Environment) text-based virtual reality with an educational theme was built.

Operating under the patronage of Hal Abelson of the Artificial Intelligence Laboratory at the Massachusetts Institute of Technology, this virtual reality was known as MicroMUSE. In December of 1993, De Anza College in Cupertino, California launched DeanzaMUSE.

All MUSE environments have a setting; a kind of conceptual framework around which the contents of the MUSE are created.

DeanzaMUSE is an exact recreation of De Anza College and the surrounding community including local schools, businesses, corporations, city government and housing areas. As well as providing a new and valuable distance learning environment, DeanzaMUSE operates as a metaphor for accessing information stored at various locations on the Internet. For example, students looking for information related to the bio-sciences need only walk into the Biology Department on the virtual campus to find links to most major biology resources on Internet. DeanzaMUSE students and faculty, operating in real-time from their computer terminals all over the Santa Clara Valley and beyond, come to explore, learn, communicate and work individually or in groups to build and maintain this rich and complex virtual campus.

This paper explores the:

The Campus Metaphor

Two factors were important in conceptualizing an environment for DeanzaMUSE: first, the need to fashion a virtual place which, by its nature, provided possibilities not available on "real life" campuses and second, the need to design an environment with sufficiently familiar landmarks so that new users felt at home in this virtual world. Since this virtual place was to be a formal extension of De Anza college with a strong educational theme and a goal of exploring new opportunities for teaching and learning, we elected to use the familiar model of the De Anza college campus and surrounding community as the environmental framework. The actual content and design of "classrooms" was left entirely up to the faculty and students using them. Within weeks, one faculty member was busy creating environments which enriched his course in contemporary Vietnamese history while others created specialized reading rooms and living areas.

Among the strongest supporters of the campus metaphor concept were librarians and research staff. Providing access to topic specific areas on Internet by placing "virtual terminals" in topic related location on the virtual campus (i.e. Astronomy in the Planetarium, Bio-Sciences in the Biology Department, etc.) offered a simple and elegant alternative to the complexities of learning to use the search tools available on Internet.

Navigation and Communication

First and foremost, DeanzaMUSE was envisioned as a highly interactive, real-time multi-user environment in which the true value of the experience was to be gained through on-line exchanges with faculty and students rather than interaction with the VR campus per se. Rather than existing for its own sake, DeanzaMUSE was designed to support, augment and facilitate teaching/learning. In support of that goal, we elected to develop our environment using the MUSE software package (rather than more sophisticated MOO programs) because of its relative ease of use, diverse and powerful communication options and simplicity of navigation methods.

In order to function effectively on the VR campus, residents must be able to move easily from place to place and communicate using methods appropriate to both the type and location of the conversation. "Places" in a MUSE environment are a complex array of logically (and sometimes not so logically) interconnected "rooms" which rely on text descriptions to evoke images of content and the surrounding area. Rooms can have a variety of attributes assigned to them and may also contain objects or exits to other rooms. MUSE uses a unique object oriented programming language which allows its members to continuously change and extend of the virtual world in real time. The center of DeanzaMUSE, and main arrival point for most visitors, is the Fountain.

De Anza Fountain(#0RLJ)

You have arrived at the De Anza College fountain, the center of the campus. To the West is the stately A. Robert DeHart Learning Center which houses the college library and teaching resource center. To the South is the Administration building and to the North an expansive outdoor dining area shaded by sycamore trees. A walkway to the East leads to the L Quad plaza and fountain.

Obvious exits: L <Q>uad <O>utdoor Dining <Lib>rary


For the most part, navigation in DeanzaMUSE is compass point accurate to the "real life" campus. To enter the Administration building, for example, residents can simply type S for south or the name, or abbreviation, of the place they wish to go.

In this example, typing S would move the resident into the Administration building where s/he would see:

Administration Building Lobby(#28Rv)

A spacious and inviting structure with the most amazing water sculpture you have ever seen. Students come here to register and pay for classes; the offices of the President, Dr. Martha Kanter and her Vice Presidents are also in this building. To the North is the De Anza Fountain. A hallway heading East leads to the Forum Building.

Obvious exits: <Reg>istration <Pr>esident <E>ast Exit


Students of all ages and nationalities pass by. The water sculpture emits another line of fascinating abstractions.

"High end" speech recognition systems provide a uniquely effective solution for removing the barriers to this virtual environment likely to be experienced by students whose physical disabilities limit their ability to access the keyboard. We are currently exploring several recognition packages from Dragon Systems in Cambridge, Massachusetts.


Essential to the use of Dragon Systems, or any speech recognition systems to be used in such an environment, is its ability to easily generate voice macros which execute a series of keystrokes associated with a specific utterance. Using the Fountain and Administration building as a simple example, our user might have the option of saying either south or administration to move to the new location. The recognition system would type the utterance to the screen and follow it with a <return> in order to execute the instruction.

For transport to destinations not in the immediate area, more sophisticated options are available. Every location in a MUSE environment has a unique address. In the previous example for instance, the address of Administration shows as #28. Using the MUSE "teleport" command, a series of natural, easy to remember voice macros can be created for fast navigation around DeanzaMUSE. Several which currently exist are: go to the library, go to the fountain, go to the Advanced Technology Center, take me home. The length and complexity of these utterances must be carefully managed so as not to overwhelm the memories of either the user or of the speech recognition system.


DeanzaMUSE offers a wide variety of methods for communicating with other residents within the MUSE or at other locations on the larger Internet. Providing these same resources to persons accessing DeanzaMUSE with speech recognition has proven to be a relatively simple task. Many of the communication tools (say, look, who) required no special modifications to the recognition system. Others, such as page (used to speak with someone at another location on the MUSE) and whisper (used for private conversations when several people are in the same room), required creation of simple voice macros such that the utterance page Sandy, for example, displays on the screen: Page Sandy =. The resident simply dictates the content of the message with which Sandy is to be paged and sends the message by saying enter key.


It must be noted that the Pop Always option in the Dragon systems software must be turned off for speech recognition commands to operate properly in a MUSE environment. It may also be necessary to adjust the Compatibility Delay setting in the recognition software, particularly when accessing remote sites via dial-up modems.

Obviously, a great many voice macros must be created and a fair amount of "fine tuning" invested in order to develop a truly intuitive and transparent voice interface to MUSE environments.

Our initial work suggests that such an interface can be used to access both MUSE and Internet resources (gopher, ftp, telnet, email) with great efficiency.

As our educational systems look for alternative instructional delivery methods to meet increasing demands for expanded instructional resources, improved services and reduced operating costs, virtual classrooms, in one form or another, may soon become a part of every student's life. By opening a window on the world outside our local communities, such classrooms may broaden our social, cultural and ethnic perspectives in new and unexpected ways. For students with significant physical disabilities, virtual classrooms may provide a "level playing field" in which ability rather than disability, personality rather than appearance, are the chief factors contributing to academic success.

DeanzaMUSE is available over Internet to educators and other professionals interested in the instructional possibilities of virtual classrooms. If you would like to request admission to DeanzaMUSE or would simply like to learn more about the project, please send email to: cbrown@netcom.com 

Go to previous article 
Go to next article 
Return to the 1994 VR Table of Contents 
Return to the Table of Proceedings 

Reprinted with author(s) permission. Author(s) retain copyright.