MANAGING THE EVOLUTION OF A VIRTUAL SCHOOL IN THE BLACKSBURG ELECTRONIC VILLAGE Andrew M. Cohill Roger W. Ehrich Deborah Hix Andrea L. Kavanaugh Herman G. Bartlett* Virginia Tech Blacksburg, Virginia 24061 *Montgomery County Schools Christiansburg, Virginia 24073 Prepared for HCIC, February 15, 1995 Revised March 2, 1995 Contact: R. W. Ehrich Department of Computer Science Virginia Tech (703) 231-5420 IN THE BEGINNING... In April, 1993, Joe Wiencko, who was then Project Manager of the Blacksburg Electronic Village Project in Blacksburg, Virginia, wrote a document describing the aspirations and rationale for an aggressive plan to network the Blacksburg community. He wrote, The goal of the project is to enhance the quality of people's lives by electronically linking the residents of the community to each other, to worldwide networks, and to information resources in new and creative ways. The entire community of Blacksburg is being used as a real-life laboratory to develop a prototype residential street plan for the country-wide data superhighway being discussed as a high priority on the national agenda. The project is being conducted so that its most successful aspects can rapidly be replicated in future electronic villages in the state of Virginia and elsewhere in the United States. The prototype in Blacksburg will exemplify four characteristics essential to a successful electronic village: 1. including an entire community to achieve a critical mass of users, 2. focusing on interactions between people rather than on particular technologies 3. providing applications tailored for each type of user, and 4. implementing the project on a timely basis, so that community networking becomes a fundamental consideration in the vision and planning of the nationwide networking infrastructure. The first Blacksburg Electronic Village (BEV) account was issued on October 25, 1993, free of charge, as it is today. As of February, 1995, 13,000 ip addresses have been issued by BEV and by Virginia Tech, 2,000 of these through the BEV office, and growth is currently at a rate of roughly 50 accounts per week. The coverage area for BEV is generally considered to be Montgomery County, a largely rural county in southwestern Virginia with a land area of 390 square miles and population of roughly 74,000, of which some 22,000 are students at Virginia Tech in Blacksburg. From this one estimates that BEV already connects about 18% of the residents of Montgomery County. The conceptual origin of BEV goes back several years further as a larger consequence of the desire of Virginia Tech to extend its computing and networking infrastructure into the community so that its information and computing resources could be made more accessible. By the late 1980s, there were well over 10,000 personal computers and workstations on the Virginia Tech campus, and the College of Engineering and the Computer Science Department were requiring their students to purchase PCs and workstations. That contributed to the need for connectivity to make central facilities accessible to all of these systems regardless of their location. Initial feasibility studies for the Blacksburg Electronic Village began in 1991, and by April, 1991 BEV had obtained verbal agreement of C&P Telephone to participate in the planning. BEV planners were intent on an Internet networking model rather than a freenet/bulletin board model because of its bandwidth and its scalability. Since by this time a considerable Internet-based infrastructure was already in place on the Virginia Tech Campus, it was convenient to extend this to the community. The first public discussion of an electronic village appeared January 21, 1992 in an article in the Richmond News Leader titled, Project Envisions Blacksburg as an "Electronic Village," following a press conference at which C&P Telephone publicly declared itself to be a BEV planning partner. At another press conference a year later, C&P Telephone, the University, and the Town of Blacksburg announced that BEV would indeed be implemented. Crucial to that announcement was the agreement of C&P, now Bell Atlantic, to invest $6 million to upgrade its switching equipment and fiber network and to provide full rate Ethernet access with 10Base T technology to roughly 600 apartments in five high density apartment complexes within the town. These complexes now provide Ethernet connections either as an amenity or for nominal fees up to $30 per month. Surprisingly, the entire enterprise has been planned and implemented without any paper agreements among the partners. In the meantime, network planning is continuing. A 500 home subdivision has begun the engineering design work for an all-fiber neighborhood network, and retirement communities and business complexes are making plans for Ethernet connectivity. Current connection cost for an apartment complex is $425 per month for a T1 line, with an option to have Bell Atlantic do complete installation to apartments for $125 each. There is much discussion of further growth of connectivity, but here the county is at a disadvantage since its 74,000 residents represent a relatively small and physically dispersed connectivity market compared with major cities. BEV clients without Ethernet currently contract with the University's Communication and Network Services for the use of its 144 port 14.4Kb modem pool at a rate of $8.60 per month. Others who wish to use BEV but who lack a computer or require only infrequent access may surf at the Montgomery County Library in Blacksburg. The Library also conducts regular networking classes which are always filled to capacity. BEV itself operates a gopher/WWW server and shares the University's mail server. A visit to http://crusher.bev.net is the best way to get a sense of the information services offered by BEV. BEV currently lists a staff of 8 that covers administration, research, technology, school liaison, information management, system administration, and office management, and it has a physical presence 9 hours a week at a modest Main Street office in the Museum of Natural History. Would-be users register there in person and bring with them an unformatted diskette that is exchanged for one with a software distribution for their computer. This includes a slip client and a set of network clients such as popmail, WWW, gopher, ftp, and newsreader. THE USERS' VIEW Montgomery County, Virginia has very diverse demographics, including at one extreme a nucleus of computer-literate, network-trained individuals typically associated with the University, all the way to rural residents without telephones who have only a limited sense of connection to an information-based society. Recently Blacksburg has begun to attract retirees across the US as its reputation is propagated by the news media. Adding to the complexity is religious fundamentalism which some perceive as threatened by free information access. BEV, of course, views itself as facilitator for communications and for information access and as an information provider. BEV has declined to act as publisher for individual county residents or businesses, and those who seek to publish on the World Wide Web are referred to private information providers, such as BizNet Technologies, which was privately created to fill this need. In its infancy BEV grew slowly, drawing first from those with computer expertise while local information resources were created and embellished. Almost immediately warning flags flew as the BEV staff grappled with the difficulties encountered by new users as they attempted to make their slip scripts and modems work with the servers and the model pool. Among the earliest registrants were influential users from town and county government; the BEV staff struggled to get their systems on line at time when they themselves had minimal experience with the myriad of technical issues associated with the new operation. The library facility opened amid much fanfare on January 15, 1994, and was an almost instantaneous success. Meanwhile the Town of Blacksburg struggled to move the information resources in its telephone-based town hotline to the WWW to add both convenience and new reasons for town residents to join BEV. In its current state, BEV is a dynamic and intimate relationship between a physical community and its electronic counterpart. BEV participants are learning new ways to conduct daily affairs, and the information and service providers are responding in kind with new and innovative offerings. People can participate in town government without going to town hall. They can more easily play a role in the county schools, or even scan a restaurant menu or find out what is on sale at the grocery store. Beginning in the spring of 1995, the Blacksburg town government will begin offering enhanced information services that will include the ability to conduct business with the town via the network. Services will include automobile registration and parking sticker fee payments; electronic payment of water, sewer, and garbage bills; electronic filing of complaints to the police department; and the processing of various other permits and paperwork electronically. In the private sector, a local grocery store chain is opening a full-service grocery store on the net, and the software as of February, 1995 is in beta test. Offering both curb pickup and home delivery of groceries, this service suggests a new chapter in retailing that looks much like the kind of personal relationship consumers had with grocery stores fifty years ago, the last time home delivery was common. The shopping software will support multiple shopping lists and standing orders, providing customers with a unprecedented level of control over how and when they shop. Offering this service will give local retailers an edge in customer service that will allow them to compete better with nationwide chains that have a pricing advantage because of volume purchasing agreements. Ultimately we must wait to see what the long-term social effects of these innovations will be, and how they will affect the relationships among residents, businesses, government, and education. There is a comforting democracy about BEV, in that a user can select the desired services and information or tune out completely. However, at least two aspects are not completely democratic. First, general publishing such as on the WWW, while part of the overall capability, is not free to everybody. As long as that is so, those who publish information on the network as the representatives of public agencies have a responsibility to ensure accuracy, completeness, and neutrality. The second aspect is deeper and more difficult to deal with. Since BEV is technology intensive, many community residents feel either suspicious toward or even disenfranchised by the network. Even a young child can learn to put a quarter into a vending machine to obtain a newspaper. In reality, surfing the web at the public library is not much more difficult, but many who have not done so believe that it is. This sets apart county socioeconomic groups and creates a problem that can be solved only by education. One is tempted to attack that problem both through the physical educational institutions and through the community network that created it. We have mentioned education last, not because it is less important, but rather because it is the central topic of this paper. It has two aspects...education about the BEV technology per se, and education as a general community agenda. As we will see these are not separate issues at all but rather part of the enormous social realignment that is taking place in Montgomery County. We have learned that most of the resources for developing a community network must be directed toward community education (Carroll and Cohill, 1994). What we failed to anticipate was the massive education effort that would be required and why. For most of us, our experience with networks over the past ten years or so has been in one or two contexts: the largely anonymous, geographically diverse Internet community, and the usually hierarchical and autocratic local organization (a university, a corporate employer) that has provided the local connection. As part of a homogeneous organization, use has been dictated by policy and protocol with little opportunity (or need, perhaps) for the kind of continuous needs-education cycle that has been vital to promoting the use of the network in Blacksburg. The problem of education has been difficult and challenging. For the next twenty years of so, those involved with the deployment of community networks will be faced with similar problems: training an adult population as well as public and business organizations to acquire new technical and social skills. To compound the difficulties, these new skills have no comparable model or analogue in the existing world. There is simply no precedent in human history for the ability to directly address hundreds, thousands, or even millions of people in a few seconds from your home. Interestingly enough, in Blacksburg, this capability must be explained and demonstrated time and again before users may be convinced of its value in their daily affairs. Somewhat counterintuitively, we have found that people in our community are not particularly impressed with the vast resources of the Internet. What seems to interest people most is better access to local, real, substantial services, rather than to the nearly infinite reaches of cyberspace. But even this enthusiasm triggers a massive education effort: merchants and business people must be taught how to package, sell, and service their merchandise on the net. Local governments must be educated about the net and why it may be a good thing to give citizens better and more frequent access to government information. Local civic and volunteer groups can leverage the tools of the net to good effect, but the often casual organizational nature of these groups makes educating them difficult. Network authoring on the WWW is an acquired art, 100% Madison Avenue, over and over again. It takes a long time for designers to understand how to go about it. A web document is a human-computer interface to the information world, and all of the understanding of the design process that we have acquired in the past two decades needs to be applied to the web. As educators, whether at the University, at a public school, or at the BEV office, we are always cajoling, selling, promoting ideas, promoting our credibility. Casual web users are easily frustrated because they can't find what they are looking for, and when that happens they may become ex-users. We have to understand not only what users are looking for but what they need to know and lead the way through big beautiful hyperlinks that shout, click me! ENGAGING THE SCHOOLS Almost from BEV's inception, discussions have been ongoing with the County schools to provide uniform access across the county. One BEV staff member was designated in mid-1993 as school liaison, to participate in discussions with the schools, and to help individuals within the school system deal with networking issues. The schools view networking as a means to provide a much more meaningful educational experience to the students by bringing world information and current events to the students and by giving them a way to participate collaboratively with peers and educators throughout the world. It is also a way to provide students and families across the County with equal educational opportunity and to reduce disparities among schools, something that has been difficult to achieve given its rural-urban nature. The school district seeks to explore a variety of network services, including cost-effective videoconferencing, to bring educational opportunities to all students, particularly to those who are most isolated and geographically remote. Networking provides a unique opportunity for students to extend themselves and to gain some sense of membership in a world much larger than that they would perhaps otherwise experience. Unfortunately, the realities of public education sometimes shed a harsh light on one's aspirations. The Montgomery County school system consists of 19 schools with 670 teachers and professionals, educating roughly 4,400 elementary, 2,100 middle school, and 2,500 high school students at a projected cost in 1995-1996 of $49 million. Included are 2 technicians and their supervisor who are responsible for all electronic technology, including networking and computing. Two more technology staff members are included in the current budget request. In general, classrooms do not have telephones, although perhaps 40% of the rooms have computers with dialup access to the Virginia Public Education Network (VaPEN) via shared phone line or local area network. Roughly 60% of the teachers use VaPEN for communication, lesson planning, and for its information resources. This significant adoption of VaPEN is in large part the result of a concerted effort by one of the teachers who was at one time the administrator of the Radford node. He coordinated a significant educational program that led to its general adoption by the teachers. However, the VaPEN modem pool is restricted to 1 hour of use daily per user, and at high load times, no user is guaranteed access either by network or by modem pool. With the exception of a few targeted programs, such as business, writing to read, and keyboarding, few computers are used in regular instruction. Most computers are a mixture of Apple IIe vintage machines, Windowsless PCs, and low-end Macintoshes. With the exception of these targeted programs, computers are used primarily for school administration. Against this background, the school system made an initial 1995-1996 request for a technology budget of $1.6 million. Of that $403,000 remained after negotiations with the school board. Even so, the total budget request to the County reflects a $3.3 million increase of 16.7% over the previous year. Should the technology budget survive intact, the success is moderated by the realization that if invested only in computers at $2,000 each, the entire technology budget would provide less than a third of the professional staff with a single computer. Unfortunately this is rather typical for a rural US school. In September, 1994 the College of Education and the Computer Science Department at Virginia Tech were awarded a planning grant of $100,000 from the National Science Foundation to work with the county schools to plan the future of network-based education and to plan for a network-based virtual school. In concept, a virtual school is an unbounded educational environment with no walls, no halls, no bells, where (virtual) collaborative classrooms encompass the entire community and exploit connections among diverse educational resources - schools, libraries, homes, businesses, government, local and global networks, and individuals. We are working toward an educational system that extends beyond the physical school into the community, in which both parents and their children continue to communicate, to learn, and to collaborate in a rich network environment. This program is undertaken in collaboration with educational material providers Busch Entertainment and Scholastic Network with additional networking support from Bell Atlantic. In our first year we have initiated a pilot program to create an emergent virtual school comprised of teachers and classes in two elementary, two middle, and two high schools, to each of which Bell Atlantic has provided a T1 communication connection. A number of research projects were either planned or under way with several individual teachers at several of these schools, but modest equipment purchases enabled by NSF came at a very crucial stage of these projects. This equipment included 6 Macintosh 6100 systems for connection to the Ethernet in the classrooms plus 7 used 80386 systems with modems to be used by teachers at home to prepare plans and to explore. 12 teachers, among them the best in the school system, were selected to participate in a series of workshops intended first of all to familiarize them with network resources and services so that they could participate in advanced planning with an informed perspective. The intent is that these participating teachers will subsequently become the mentors for the next group, and that model seems to be working. Then came a period of time for the teachers to work with the technology with intensive one-on-one assistance from Virginia tech faculty, students, and staff while the T1 lines were connected within the schools. At first progress was slow, and then suddenly came a burst of intense activity when the teachers overcame hardware and software problems and a variety of problems with accounts, passwords, and physical connectivity. Soon the first few teachers mastered html and began publishing school information on the WWW, lessons for their students, and student work, all of this evenings and weekends. Teachers began to take their classes to the networked New Media Center at Virginia Tech (a public laboratory, largely supported by Apple Computer), where they could work on their network-based lessons. Two teachers are working with their students on a networked multimedia magazine. On teachers' workdays, mini courses are now offered on computing, networking, and html. By the end of February the first school will have a formal presence on the WWW, along with the central administration, and a number of others will follow. Again, this effort is being managed by motivated teachers evenings and on weekends. In fact, the success of the program so far has surprised and pleased both grantees and teachers. MANAGING THE EVOLUTION Our planning grant is an intense and carefully managed program, although it is difficult to organize such a physically dispersed group of people working on schedules that make them generally inaccessible during the day. Despite our elation with our successes, we have a number of observations that bear directly upon plans for the future to use computer networking to change the quality and effectiveness of education. First and foremost, the mere presence in the classroom of a networked computer has little effect upon educational process and outcome. In fact, most teachers will use a new computer much as anyone else would for educational administration and for the conduct of professional activities. Only the most creative and accomplished teachers have innate ability to map the peculiar resources of the Internet into a directed series of lessons that achieve the educational objectives. The network is anarchistic and timeless and gives up its secrets only to those who approach it with methodology, patience, determination, and a sense of adventure. Teachers must be allocated time to explore network resources, to reflect, and to synthesize new approaches to their educational program. They must be given technical assistance at a personal level to help them over come both technical and conceptual barriers, and they need mentoring and encouragement from colleagues who have already mastered the new skills. Teachers are problem oriented and generally put off by the necessity of dealing with technological details that seem irrelevant to their goals. The bad human factors that seem to characterize school computing discourage teachers from serious involvement with network-based instruction. Computers must boot when turned on, and there must be support personnel to tend immediately to those that don't. A malfunctioning mouse must not be allowed to sabotage the week's carefully laid lesson plans. There are undeniably documents on the Internet that are not suitable for K12 students. However, these risks have received so much attention in the public media that parents and teachers who don't understand the technology may over-react. The risks need to be carefully discussed, and an approach should be adopted that combines student supervision and direction with calm, matter-of-fact discussion with the parents. Parents must be made aware not only of the risks but of the very significant benefits to their children. As a group, the parents suffer more from culture shock than either the teachers or their students. They tend not to understand how poor school computing facilities are and how that affects the possibilities for their children. They don't sense the direct connection between their tax bill and their children's future...perhaps they don't have the confidence that their tax dollar will be spent wisely. The parents have a difficult time gaining access to the information that will help them understand their schools, the information that they need to make the best decisions. Again, public education involving publication of relevant information and open discussion seem to be important to overcoming mistrust and misunderstanding. Educational change depends heavily upon the presence of visionary leaders who understand the risks and the gains and who can mobilize support by persuasiveness through tireless public appearance and demonstration. Clearly the schools desperately need financial assistance, though we argue that this is essential but not sufficient. For the sake of argument, let us adopt a naive and simplistic model of how the transformation might be achieved. From the information presented earlier, at $2,000 per computer, a ratio of one computer to four students and faculty members requires an investment of $5 million. 1 month (20) days of training per staff member would add $3/4 million to that total in released time alone. Then there are the continuing costs of network connections, hardware and software upgrades, and support personnel, which could easily be $1 million per year. In a society as wealthy as ours this goal is easily achievable, provided there is a unity and resolve in the community will that this is necessary for the future of our children. Other investigators (Hodas, 1993) have studied problems associated with technological change in school organizations and have concluded that because of the complexity of the system of constraints that must be satisfied by a school system and its staff, real educational change associated with the introduction of new technology will be difficult to achieve. A conclusion reached by a number of those involved with the planning project is that to introduce educational networking properly, it will be necessary to create a separate school, perhaps a so-called exemplary school. This school would adopt the best possible curriculum, the best teachers, and the best technology, and then propagate the new culture outward into the entire school system. Some have observed that a curriculum such as an International Baccalaureate curriculum seems very consistent with networking technology, and this is now a topic of serious discussion. Educational change in the schools will be difficult without corresponding changes within the families and the community. One of the best potential agents of change in our community is the electronic village itself, but it will take massive education to turn it into an effective agent of change. Ways to draw the parents into closer collaborative relationships with the schools and with government must be explored. If we are successful, people will no longer have to move physically to acquire information...it will come to them. We will have our virtual schools. A SHORTER RANGE AGENDA One doesn't have to spend a lot of time with teachers to understand how incredibly sensitive their attitudes toward computing are to the human factors of their environment. As HCI professionals we have long been surprised how important long range decisions are being made without any focus of concern on the human factors of the school computing environment. For example, the Internet is full of different examples of FTP clients with different interfaces that range from textual to graphic. Some that are wonderfully graphic may crash repeatedly or have no recovery mechanism for frequent server time-outs. We wonder how one can reach substantive decisions about the viability of networking environments in the schools without first optimizing those environments. In general teachers have to put up with human factors we wouldn't tolerate on our own home computers. Beyond that, network services are evolving in a very uncontrolled way. It happens frequently that a small group of very talented networking specialists creates a new client or network service. If it generates sufficient excitement it becomes immediately popular regardless of its usability. Being shareware, it may never undergo formal usability testing and iterative refinement, and before long this untested interface may become an international standard. To cite an example, we recently performed a formative evaluation of a popular video conferencing program from Cornell University called CU-See Me. We find the program conceptually fascinating and performed this study because it was to be the basis later for a formal study of its use in 5th grade geography instruction. Multi-party video conferencing poses some interesting usability problems that suggest themselves quickly when one observes a typical group of a half dozen people testing the software, tapping on the microphone, and typing, did you hear me? Visually it suggests the gorilla's view of the zoo. One problem is that the visual cues are all mixed up. A speaker addressing one party may not be seen to move his lips because of the low frame rate, and the darkening speaker icon may be missed because of its position on the display. On the other hand another participant may think that he has been addressed instead. Generally, the participants spend their time scanning the video windows and associated interfaces for activity cues, to the extent that they have difficulty participating in the conference. The formative evaluation was actually addressed at novice users in a conference with two or three participants. The results suggested that users at all computing expertise levels had difficulty getting started with the interface because of a number of design features of the interface. Although one subject walked out of the experiment, the rest seemed to learn the interface quickly because it is not that extensive. To return to the educational networking agenda, the Internet services available in the popular networking suite make substantial contributions to private mail communications, published mail communications, and to digital libraries. Recently input components have been added to WWW html scripts to permit a browser to return a form, but the basic WWW and gopher paradigm is an asynchronous one in which server processes don't wait around for responses from the clients they serve. In education, active learning is enhanced through processes of collaboration in which learners collaborate with software agents, with each other, or with groups, possibly solving complex graphical problems through a shared workspace and world model. Although there are several possible mechanisms by means of which collaboration (which is known as synchronous collaboration) could be supported by the WWW, there is little work to support education aside from some feasibility studies performed at some of the networking laboratories around the country. It is out belief that network-based collaboration is a fundamental and necessary addition to the educational networking suite, and we will be pursuing it from our perspective of human factors and the social aspects of learning. REFERENCES Carroll, John M. and Cohill, Andrew M., "Managing the Evolution of the Blacksburg Electronic Village," November 11, 1994, unpublished manuscript. Hodas, Steven, "Technology Refusal and the Organizational Culture of Schools," Educational Policy Analysis Archives 1, No. 10, September 14, 1993, available through the listserver edployar@asuvm.inre.asu.edu. The authors gratefully acknowledge the assistance of Mr. John Schorger with the time sequence associated with the development of the Blacksburg Electronic Village, which is part of his dissertation research.