The Present
Many believe a revolution is taking place in education in the way people learn and the way instruction is given. The education community has been hearing of reforms and revolutions for the past few decades, but most of them have been nonexistent or without any long-term merit or real value. Some believe the method of an instructor lecturing while students listen and "absorb" is really the only viable way to teach or learn. About two decades ago, when personal computers started to become affordable, many thought that computers would revolutionize education, that computer-based teaching and learning would become the savior of education and the solution to falling test scores. This has never really happened. Over the past two decades, many teachers have successfully prepared students, some with computers in the classroom and some without. Teachers could avoid computers, either because they chose not to learn how to use them or because they had none in their classroom or school to use. Teachers entering the profession have not been required to understand computational technology in order to graduate from college.
The Internet has been in existence for almost two decades and began to extend into schools about 15 years ago, first into universities and then into K-12. Did the Internet revolutionize education? Well, not exactly. It did provide an opportunity to expand learning options for teachers and students who were fortunate enough to have Internet access, a few computers, and appropriate guidance on usage. Often this took place in only one classroom and only one school within a system and did not become systemic throughout the school. There are many factors affecting this slow implementation of computing and communication technology in schools, including administrations with no knowledge of its value or no willingness to realign school budgets to include computational technology; insufficient in-service professional development programs for teachers; a lack of specific curriculum benefits or of resources for teachers to use in their courses; and deficient preservice preparation of teachers in technology or computation.
Why do some of us believe there is now a revolution taking place that cannot be ignored by educators or administrators? In November 1993, the National Center for Supercomputing Applications (NCSA) released Mosaic, the first World Wide Web browser for all three computing platforms (UNIX, PC, and Macintosh). The Internet had become the World Wide Web, and now Mosaic allowed anyone who knew the basics about using a computer and a mouse to go out onto the Web and easily and quickly locate multimedia information. Suddenly text, images, sound, and video could come to the desktop and be used by students from the kindergarten through Ph.D. levels. Files rapidly became hypermedia text pages, and nascent searching and information integration tools became easily accessible. So what is the difference between the past two decades of computers and Internet access and the present, since even now WWW access requires computers and an Internet connection? The difference is that the Web represents Information, and information cannot be disregarded the way that computers can be ignored. Teachers cannot choose to ignore or have their students omit available information on any subject when the goal is for them to learn. A revolution is taking place in education.
The Future
While the first impetus for schools to connect to the Web has been simply to access information and this incentive has made a great deal possible collaboration around numerous kinds of educational activities will become a primary motivation for connectivity among schools, teachers, parents, and students. When students, teachers, parents, administrators, and even legislators can communicate via the Internet and begin to collaborate electronically on issues, the traditional educational process may see a fundamental transformation, with decisions about a student's learning being resolved in new and hopefully better ways.
Traditionally, schools and classrooms have tended to discourage many forms of collaboration. In the archetypical system, decisions related to course content and delivery arrive top down from the administration, teachers spend all but a small portion of their day confined to a room, telephones are not easily available to teachers in most situations, students are rewarded for not talking and for working independently, and communication between teacher and parents is scheduled once or twice a year at most. Teachers who value communication with parents and students generally must do it on their own time from home. Students who want to or have to collaborate must often do it outside of the scheduled "work day."
In contrast, success in graduate school, business, or life in general relies on collaboration and teamwork. The traditional education system, being more evolutionary than revolutionary, is unlikely to transform itself any time soon into an environment that teaches and encourages collaboration as a part of learning; emerging technologies, however, can catalyze this change much sooner than it would happen otherwise.
electronic mail and "surfing" for information with a browser, such as Mosaic, involve communicating with others and locating information, but the real power of the Web will come from people being better able to accomplish their "work," regardless of its focus. Communication via e-mail is a form of asynchronous collaboration. People enter messages that are sent to other people who read them, and perhaps respond, at a later point in time. Bulletin boards, such as the nearly ubiquitous Usenet newsgroups, are another form of asynchronous communication, to an open audience. Listserves based on e-mail are a very similar form of asynchronous, group communication. Desktop video teleconferencing, on the other hand, allows real-time, synchronous collaboration, although bandwidth is consumed quickly with this technology. Chat sessions are a text-based version of synchronous collaboration, as are Multi-User Dungeons (MUDs), Object-Oriented MUDs (MOOs), or Multi-User Shared Hallucinations (MUSHes), where a number of people participate simultaneously in a shared conversation or activity. The real power of digital technologies is not yet tapped; other applications on the desktop for doing exciting and robust synchronous and asynchronous collaboration are just now emerging.
Although a number of schools in the nation have invested in traditional video teleconferencing classrooms, and although some students have probably benefited from these facilities, this technological investment is not expected to be a cost effective or a pedagogically valuable way to extend distance learning to the nation's students. What we do expect to flourish and grow are asynchronous and synchronous collaboration applications between desktop computers, some of which may be hooked to a projection system so that many people in the same location can participate. All the desktops will be connected through the Internet. An instructor can sit at a desktop workstation and communicate with a "classroom" of students, each of whom has a workstation in front of him and a connection to the Internet. The students no longer have to be in the same room; they can interact with other students and teachers in different locations.
Emerging Technology in the Service of Collaboration
An example of the kind of collaborative software that we have in mind is growing out of a software project of the National Center for Supercomputing Applications. A few years ago, NCSA developed a system called NCSA Collage intended to help computational researchers share data from simulations, images generated from that data, text, drawings, diagrams, animation sequences, and screen shots from other analysis software with remotely located colleagues in a real-time, synchronous session. This software addressed a continuing need of research teams scattered across the globe. Combined with a conference phone call or a chat session, Collage allowed researchers to carry on a virtual meeting, with many of the resources of their digital tools available to the meeting. Researchers soon began asking us to find a way to enable them to easily locate relevant papers or journal preprints on the Net and bring them into their Collage discussion sessions. Looking for an answer led us to the early existing WWW software. We saw that with improvements in their ease of use, additional functionality, and a cross-platform implementation, these graphical, hypertext-based Web browsers could be of great help in addressing our problem. This led to the development of NCSA Mosaic.
The Mosaic story is by now well known. Soon after its introduction, NCSA Mosaic caused a 100,000-fold increase in WWW NSFnet backbone traffic and brought in millions of users, thus creating a market for the dozens of commercial browsers that subsequently have been introduced. Since then, the commercial market has taken off phenomenally. Spyglass, Netscape, and Microsoft all have produced widely used browsers and Web technology, and scores of companies have jumped into the Web market. Spyglass alone has created a licensing family of 36 companies with over 80 products. Netscape was created by a half dozen NCSA-trained programmers and is currently a market leader. The Microsoft Internet Explorer is a licensed derivative product from the University of Illinois Urbana-Champaign and Spyglass that is available to anyone with a Windows 95 interface a potential market projected to number in the millions within one year. At last count, there were over 40 different browsers available on the Web. On the server side, NCSA WWW server software is used on roughly 50 percent of the installed base of WWW servers on the open Internet and about 70 percent of those in the .EDU domain (which consists of educational institutions). In addition, commercialization of server software is proceeding at an astounding rate.
With the explosion of commercial effort in Web-based technologies, it is clear that the Web is here to stay. In one form or another (and the exact form is both extremely important and virtually impossible to predict), the growth of access to TCP/IP-based networks the Internet as we generally know it and the availability of Web-based software and services, will only increase, driven by a vision of huge market potential. At the same time, the cost of access and services in this extraordinarily competitive sector will only decrease. The rise of new technologies and the constantly decreasing cost of network bandwidth and processor power will help to insure that the Web is more than just a short-term trend. The next generation of collaborative tools needs to be Web-based.
It is also important to recognize that the unique culture of the Internet, based historically in the cooperative environments of the academic and research communities, will not disappear in a more commercialized Internet. While there is a continuing desire on the part of companies to lock users and developers into proprietary formats and programming interfaces, there is also a need for open systems that can be extended and modified in various ways and that can cross the boundaries of different operating systems and applications. This quality of interoperability has been one of the fundamental advantages of Web-based applications.
The research and education communities also have established a tradition of placing their software in the public domain or, alternately, making it "free with copyright" along the model of the NCSA Mosaic software. Under this approach, individuals can use software for free, while the authors still retain the rights for commercialization. The dynamics of the Web have transported this idea into an increasing number of commercial software sectors, from browsers to programming languages. This makes new technology widely available almost instantaneously and gives users and developers a rapidly expanding spectrum of possibilities for tool development. On the other hand, some have dubbed this the "heroin approach" to software adoption: users get "hooked" on free tools, then later pay to get the latest versions, full support, or software that is no longer available for free. It is too soon to tell how these proprietary issues will turn out in the long term, but the educational community is often one of the main beneficiaries of free or low-cost software.
Integration with the Web
NCSA is now working to bring to the Web the kind of technology developed in Collage and to integrate synchronous collaboration with asynchronous methods. The Web provides a set of standards for handling a wide variety of media types and an installed base of browser software and server environments that can be integrated fruitfully into a framework for using collaboration methods in education. The Web has become a condensation point for virtually every advanced technology developed over the last two decades by the computational science and computer science communities. All of these technologies are streaming onto the Web at fantastic rates, each bringing its own revolution in how we conceptualize the future digital communication infrastructure. Many advanced and disparate technologies are being integrated into the global system, including object-oriented frameworks for application environments and databases; state-of-the-art security methods implemented at the connection, document, compiler, and interpreter levels; efficient distributed three-dimensional modeling and scientific visualization schemes; robust and truly persistent global naming and location methods; semantic-based object description and search techniques; and our focus in this paper workgroup and workflow software support environments in real and non-real time. The following examples show what these technological advancements can mean for learning environments.
Remotely controlled Mosaics represent a collaborative capability that could make distance teaching a much easier and more meaningful way to learn. An instructor who knew the Internet Protocol address of each student's machine and had control privileges could regulate the presentation being delivered to the students and, via the WWW and a browser, could present multimedia material and information just as he or she would in a traditional classroom with all the students present. The instructor could also release control of the students' machines to enable them to work on exercises, look for information, or communicate with other, remotely located students involved in a group project. The instructor could regain control at a future time to "discuss" an item with the entire class. In addition to presenting previously prepared information to the student machines, the instructor could give new information to all students by preparing it in advance for delivery, by typing it in, or by using voice and, eventually, real-time digital video.
In this type of classroom, anything on the Web would become material for shared, real-time presentation and discussion. Adding a simple capability to draw on the browser window would provide the users with "white board" interaction. Any changes that the instructor or a student made in this window could be seen by all the participants in the virtual classroom, and it would become easy to point out across the Net the important features of an article, a map, an image, or a 3-D drawing.
By generalizing the capability that allows one user to control or drive the software running on another user's machine when both are connected to the Internet, it becomes possible to bring any collaboration-capable software tool into the interaction. In this way, specialized educational software could be integrated into the virtual classroom; an instructor could demonstrate to a remotely located student how to execute an application. The ability to run simulations on remote or local machines is another application that is very important in teaching computational science and indeed could be useful in teaching chemistry, physics, and other high school science subjects. In the past, this application has been accomplished not with a browser, but with a series of File Transfer Protocols (FTPs), telnets, or other operations that allowed connectivity to another machine. Providing interaction with remote machines through a straightforward interface like a Web browser would make this capability available to a much larger audience.
Building these types of applications for the Web requires much more power in the Web browser than we currently have. In fact, the idea of a browser needs to give way to the more encompassing idea of a Web environment that is flexible and dynamic and can be customized to the needs of the moment. Java, a new programming language from Sun Microsystems, opens up these possibilities by allowing programs to be sent over the Web in a secure fashion. Programs that are executed locally when they arrive at the user's desktop are an example of "mobile code." The possibilities engendered by secure mobile code on the Internet are staggering. Sun's Java is an example of a C++like language that was developed specifically to address network security considerations. The capability to send executing programs that can link dynamically with existing desktop browsers and other software means that whole applications, protocols, and interfaces can be sent over the Net with HTML documents.
NCSA is evaluating the Java framework and working with Sun to make this technology available to educational users. We are working with Sun's Hot Java Web browser, investigating the use of Java in high performance systems and applications, and developing a foundation set of software to enable advanced collaboration across the Net for all users. NCSA is also collaborating with the Open Software Foundation to evaluate Java's security features, with the University of Illinois Digital Library Initiative project to develop advanced search interfaces in Java, and with the Online Computer Library Center, Inc. to study the use of Hot Java for complex scientific documents. The use of mobile code (and Java is only one example of mobile code systems) very likely will revolutionize how students and teachers use the WWW, equivalent in impact to the introduction of Mosaic and the WWW a few years ago or to the advent of NCSA Telnet and FTP before that.
Habanero
Java is also the foundation for a new environment, named Habanero, that will encompass, and go beyond, the capabilities of Mosaic and the synchronous collaborative features of Collage. The collaborative engine developed in Java will allow users to share anything that can be sent over the Web (HTML, graphics, data, etc.), in addition to live sound and video. The basic Habanero engine will include a software interface that will allow any application to become collaborative by conforming to the interface. As in the above hypothetical scenario, a white board setup would allow users to share a drawing appearing on any page in the browser by sending drawing messages to the collaboration engine. Any piece of data analysis or visualization software could become part of a collaborative session by exposing part of its interface to the Habanero engine, thereby allowing users to incorporate advanced or discipline-specific applications into their collaborative sessions. In addition, any collaborative session could be recorded, reviewed later, synopsized, annotated, and distributed to those who missed the virtual meeting or classroom session, with all the media and interactions of the original session intact.
NCSA's initial efforts focused on incorporating the functionality of Collage in Habanero, which will provide a core set of collaboration tools for all researchers. Our latest efforts focus on developing shared HTML editors, audio, easy third-party software integration, and agents for meetings (which serve as your representative to collect data or notify you of the need to respond to a request).
The availability of additional bandwidth through, say, cable TV Internet access, would be of obvious value in enhancing real-time collaborations, especially those involving huge data sets, high speed visualization, and real-time video. Collaborative uses of Virtual Reality Modeling Language (VRML) would also be enhanced by the increased speed and would pave the way for the next generation of research and development of fully 3-D collaborative environments. NCSA's strategy builds on exploratory research done in the CAVE, a fully 3-D virtual environment with virtual presence of the collaborators in the form of "avatars," or rendered representations of the participants. This technology is moving toward fully digital conferencing techniques that go beyond video and encompass a completely rendered, customizable, bandwidth-sensitive, 3-D representation of each participant, in a fully 3-D environment for the conference and any accompanying data or simulation results. Bandwidth and workstation capabilities available to users can define a spectrum of these kinds of environments, from full immersion in something like the CAVE, to VRML-based 3-D environments on widely available desktop systems.
Web as Habanero: Integrating Synchronous and Asynchronous Activities
The Habanero development is part of a more encompassing effort to bring the power of high-performance computing to researchers and educators on the Web. In its initial development by Tim Berners-Lee and Robert Cailleau at CERN, the European Laboratory for Particle Physics near Geneva, Switzerland, the World Wide Web was originally envisioned as a system to organize and enhance the work of research teams in high-energy physics. The popularity of the Web as a means of information discovery and retrieval (the global electronic library metaphor), and as a foundation for electronic commerce (the electronic market metaphor), has tended to push into the background considerations about its use as a global research lab (the virtual institute or virtual classroom metaphor). NCSA is focusing on developing a Web infrastructure that would change this and provide capabilities directly targeted on globally distributed research and learning communities. An example is the development of HyperNews (a Web-based bulletin board system) and of global annotation capabilities, which can be used as a foundation for virtual research colleges (in the classic sense) or research institutes on the Web.
These kinds of Web-based virtual institutes could include most or all of the functions of a brick-and-mortar research institute. A research team or a class could form an institute by placing notes or annotations on documents, data, services, indexes, bulletin boards, discussion groups, or anything else on the Net, thereby signifying that these objects are part of the work of the institute, part of the relevant neighborhood. A reader would be notified when she entered the neighborhood of the virtual institute and would be given information on who else was visiting, what the rules of the institute were, what members' office hours were, what research projects were underway and their status, and whether any active collaborat)ons with others were being sought. In addition to participating in asynchronous collaborations via bulletin boards, e-mail, and the like, a visitor to the virtual institute could establish synchronous collaboration based on a chat session, a Collage session, a digital video teleconferencing center, a shared VRML scene, or any combination of these. She could then join an existing collaboration or inquire whether any institute members would like to start up a new one. A more simple use of this system would allow any geographically distributed research team or learning group to organize their activities with full-media support, or would allow a teacher to organize the lesson or project material of students.
An alpha version of this Neighborhoods software has been completed, and work is starting on a VRML neighborhoods application. With the VRML application, users visiting a VRML scene can "appear" in the scene via an icon and, with the aid of an integrated chat box, can communicate with other users that they can see. Future work aims toward a collaborative VRML environment, where users can run shared VRML simulations or take "group tours" through a VRML world. Integration with the Habanero framework and HyperNews will proceed over the next months.
Repositories
To provide some of this functionality, server-side capabilities must be enhanced and made available in forms as easily adoptable as current HyperText Transfer Protocol servers are. Many of the ideas discussed above were first applied to high-performance computing and communication systems, and a wealth of experience and collaborations with technology innovators has led us to establish a framework that will allow Web testbeds to be prototyped rapidly and permit numerous technologies to be investigated and integrated. This modular, flexible, and extensible framework allows various kinds of functionality, including collaboration, to be included.
NCSA is using this Repository framework, as it is called, to pursue work in many different areas: high-speed Asynchronous Transfer Mode networks for the Web; digital libraries for managing Standard General Markup Language collections and other complex, highly structured scientific documents; global naming, annotation, and collaboration systems; sites for managing and distributing valued objects with strict terms and conditions governing their use and their intellectual property rights; massive digital data manipulation and access points; object-oriented databases for Web site contents; advanced semantic-based searching of information across disciplines; and distributed site management for large public data sets. NCSA is also working to rapidly integrate those capabilities that prove themselves in the prototyping phase of our next generation of public software.
Federations of Repositories
A Repository is a framework for integrating advanced Web services. NCSA has defined a structure to enable individual Repositories to better manage their information and has gained experience from building several Repositories; our next task is to determine how Repositories interrelate. The basic framework will establish the capability of communication between Repositories, but what they communicate about and how several of them cooperate to provide larger federated services are open questions at this time.
One way in which Repositories can interrelate is by providing a search service that transcends Repository boundaries. Support for manual searching, or browsing, already exists in the WWW; here the user selects links to follow from one document to the next across the Web of related documents. Automatic searching over the Web that uses the same links between documents requires standards across servers for query language, knowledge representation, and metadata semantics (a new and very powerful way to search semantically). Automatic analysis of collections of interlinked documents can also help produce cross-collection indexes to optimize searching in fairly focused areas. Collaboration technologies and this Repository framework can also allow users to contribute to the incremental self-organization of the Web.
NCSA is currently developing a framework for educational information in Illinois called the Illinois Learning Mosaic (ILM) project. This Web-based information resource and database will support information of interest to students in kindergarten through the Ph.D level, as well as to teachers, administrators, parents, legislators, and other citizens. The ILM is being designed within the Repository framework of future Internet server software, so that any other information databases on the Internet built within this same Repository framework will automatically share data. When this work is complete, servers will be communicating and sharing with servers for the first time. This will allow anyone on the Internet to develop interrelated data bases and collections that automatically share similar data. This type of sharing between servers also forms the foundation of the annotation capabilities underlying the Neighborhoods and Habanero projects.
Workplace of the Future
NCSA has just finished the beta version of a new framework for collaboration on the World Wide Web called the NetWorkPlace. Completed for Vice President Gore's National Performance Review, this environment of asynchronous communication and collaboration tools allows distributed team members to work on the Web in a new and potentially revolutionary way. Some of the current tools include a Web-based group calendar, chatting capability, capability for threaded discussion and group meetings, writing tools, document storage and publication tools, a project management tool, a forms tool, workspace management tools, and more. New asynchronous tools are currently under development, and the synchronous tools being developed through the Habanero effort will become features of this workplace very soon. NCSA is also implementing tools and capabilities for classrooms and schools to use the new collaborative framework, and within a few months we plan to produce a framework for a "virtual school."
Conclusion
Technology is affecting education in revolutionary ways, and the momentum toward these changes is irreversible. Teachers who have begun to use the Web see this change occurring, even if they only have experience with static information-gathering and display capabilities. Most of these educators have not yet used or even seen the potential of collaborative technologies for their classroom and their school. The majority of the capabilities discussed in this paper have not yet become functional in the classroom, but they will very soon.
Universities need to become leaders in applying technology to education for learning and for collaboration. Colleges of education need to become leaders in applying computational and information technology for the K-12 community, and university administrators need to begin to chart the 21st century vision of their institutions, a journey that will include information technology and collaborative learning and teaching.
An obstacle that needs to be overcome is the view many hold that computers and Internet connectivity are "tools" for learning, and thus an increased grade point average is the only measure of value for these resources. A more important perspective is for administrators and school boards to realize that the Web represents a new environment for learning and teaching and that very soon every teacher and student will need access to the information represented on the Web in order to be competitive in their work and in their lives.
