Stephen C. Ehrmann, Ph.D.
Program Officer for Interactive Technologies
The Annenberg/CPB Projects

The author, active in national funding of U.S. applications of technology to undergraduate education since 1978, concludes that courseware developed and distributed for instructional use has not played the central and revolutionary role that he and others had expected in the late 1970s. Instead the use of technology in undergraduate education is being increased by the need for undergraduate courses to incorporate the same kinds of tools and resources that professionals use so that undergraduates can begin to learn to think like professionals. Institutional needs to maintain or increase enrollment also have offered a potent motive to use technologies such as live video and electronic mail. These changes are large in scale but almost invisible since they are the result of local action.

What can we learn from the United States' effort to enrich and extend undergraduate instruction through applications of computing, video and telecommunications?

My viewpoint is unavoidably personal and rather unusual. From 1978-85, I worked for an agency of the United States Department of Education that offered grants for innovative proposals to improve higher education. Many of the proposals sent to this agency, which is called the Fund for the Improvement of Postsecondary Education (FIPSE), had to do with computers and television. During that period, FIPSE's grants for technology increased as we received more good proposals; by its end we were spending almost 30% of our appropriation on them: almost $4 million/year. Since 1985, I have worked for a funding program called the Annenberg/CPB Project at the Corporation for Public Broadcasting, which, during the period 1981-90, invested $10 million/year in the development of video, print, and computer software for instructional use, demonstrations of new technology being applied to educational problems, and research on technology in higher education. If this essay gives you the misimpression that funders were at the center of the movement to use technology to improve university education, I apologize. I was in the center of the action that I could see from where I was sitting.

Before I go further in relating what I have seen, let me insert one note on my language. In the United States, the line between colleges and universities is blurred, but it is still meaningful. However, as Kerr did in the preceding quote, for simplicity's sake and to avoid confusing those for whom the word "college" refers to something I would call an academic division, I will use the word university when I mean an institution of higher education that offers a two- or four-year academic degree.

Back in the late 1970s and well into the 1980s, my work was guided by several assumptions that I don't believe (so much) anymore:

* that the essence of what technology could do to revolutionize education (and that indeed was what I hoped, and hope, for) was represented by courseware developed for instructional use and distributed nationally,
* that video was best used to extend the accessibility of higher education while computing was best used to improve and individualize its quality,
* that revolutionary new materials could and should be created by learning to exploit the strengths of a single new technology, and
* that technology-enabled change in higher education would emerge as the sum of actions taken by individual faculty to improve individual courses.

Perhaps you think I was pretty silly to believe all those things. Let me take you back to 1972.
A VIEW OF THE REVOLUTION, CIRCA 1972

By the early 1970s, (mainframe) computers had already become a widely discussed means of improving education, at least in the future. One volume exemplifying that mainstream view of technology was part of a series of reports on how to improve colleges and universities that was sponsored by the Carnegie Commission on Higher Education. That book was entitled The Emerging Technology: Instructional Uses of the Computer in Higher Education. In the foreword, Clark Kerr, chairman of the Commission, noted that in 1969 1200 institutions, over a third of the nation's colleges and universities, had at least one computer; that marked an increase of 22% in just one year. The volume reported that the computer received its greatest use in research (40% of applications in 1966-67) while instruction accounted for 30% of computer use. After mentioning briefly its instructional usefulness for clerical chores in libraries (e.g., monitoring circulation) and classrooms (tracking student achievement), Kerr turned to the main topic of the volume, the computer's use for "presenting information to students in an orderly fashion, periodically testing comprehension, and automatically advancing the level of instruction as students are ready for it." (pp. xv-xvi) While asserting that computers were already performing instructionally useful roles, Kerr emphasized the barriers impeding further progress, including lack of access to machines for all students (25% of all students had no access to computers for instruction, he estimated); lack of standardization of equipment; a shortage of instructional software compounded by uncertainty about whose task it is to supply such software; faculty members fear and mistrust; and lack of the kind of information needed by policy makers. We can pause to note that the barriers sound quite familiar in 1993, even though the personal computer had yet to be invented in 1972 (in fact one can find very similar barriers cited in a 1937 study of a medium that was supposed to revolutionize education: educational film ). Roger Levien, the volume's senior author, asserted that

"the next 25 years will be a universal era, in which, for an extremely broad range of applications, computers will become economical and consequently will be widely applied to tasks that now seem infeasible. For almost everyone, the computer will be accessible, both physically, because it will be close at hand in the form of a typewriter or television-like device; and intellectually, because it will become a convenient, tireless, and reliable assistant in daily tasks, both mundane and original." (p.3)

Levien asserted that the computer would have several roles, as subject ("studies of the computer can be expected to be part of the curriculum and research plans across the campus"), as tool (for faculty members, administrators, and librarians), and as agent of change increasing the accessibility of information and instruction. He also predicted that computers off-campus would be linked to campus computers, so that working adults would be able to "tap the university's fund of information and instructional skills." (pp. 8-9)

Levien seems quite prescient in what he does predict. Like most people involved in this movement for the next decade and more, however, Levien, like Kerr, assumed that the dominant instructional use of the computer would be "presenting information to students in an orderly fashion, periodically testing comprehension, and automatically advancing the level of instruction as students are ready for it." He assumed that the computer would function as a tool for faculty but not for students. We will return to that mismatch between vision and reality, and its consequences.

The other major technology that was being applied to higher education in the 1970s was television. Television made it possible for interested colleges to offer a scattering of college courses to adults whose location or, more often, schedule made it difficult for them to study. Typical users might well take one or two courses on-campus and another one or two via telecourse. Television-based courses at this time often assumed one of two forms: 1) as much as fifteen hours of prerecorded video would be received by students through public broadcasting; this represented the equivalent of on-campus lectures; 2) live classes broadcast via microwave to industrial and other sites.

Thus, in the 1970s, computing was being hailed for its potential impact on the quality of higher education (via faculty-developed, distributable courseware) and video was being promoted primarily for its promise to improve the accessibility of higher education.
DISTRIBUTABLE COURSEWARE AS PARADIGM

While at FIPSE and later at the Annenberg/CPB Project I had a hand in providing the money for a significant number of courseware development projects. I am proud of almost all those projects. In order to survive ferocious competition for funding (as many as 30-40 proposals from a wide range of institutions for every available grant) the authors of their grant proposals had to be educationally adventurous -- able to write the kind of proposal that would give a reviewer shivers, and make him glad he lived in an era where he could start a revolution. For example, one of the first projects I helped, a videodisc of biological images, was funded around 1980 and the product (in an upgraded version) is still on the market in 1993. Most of the other products were completed and used with great success, at least locally.

And yet, as the 1980s wore on, many of us were concluding that such courseware was not causing a national revolution in instruction. What was wrong?

Three years ago, a number of us active in EDUCOM's program for Educational Uses of Information Technology (EUIT) began talking together because we were disturbed by the spreading perception that computer courseware was failing to be widely used and to transform higher education. Were there families of software that were not only valuable but also viable? We defined value to mean educational usefulness (as demonstrated by prizes won or evaluation results, for example.) We said that software was viable if it were used widely enough and for long enough so that all the parties to the software's development, distribution, and use could feel reasonably contented about the return from their individual investments. This Valuable Viable Software Project team was composed of academics, publishers, and hardware vendors who had volunteered to work together to solve a problem important to all of them. Dr. Paul Morris, Director of Computing and Communications at Tufts University in Massachusetts, soon became our leader.

Without attempting to repeat our findings, here are a few of the things we found.

Like most people since Levien, our attention had initially focused on software that was developed for instructional purposes in particular courses taught at large numbers of institutions: courseware. We knew, of course, that many other types of software were in instructional use in colleges and universities, but distributable courseware was supposed to be the crown jewel, the agent of revolution. (We decided early on that courseware developed by instructors solely for their own use was not critical enough to warrant inclusion in this particular study.) However, by sharing the collective knowledge of the thirty of so academics, publishers, and others in our group, we discovered that we had heard of only a few pieces of distributable computer courseware that were both valuable and viable.

Finding so few examples of valuable, viable computer courseware was not a huge surprise; it was anxiety about that very issue that had drawn us together. We knew that the market for computer courseware had proved to be small and brittle, for several reasons.

First, higher education is smaller than the kindergarten-high school market, and the variety of ways in which one course may be taught is greater. So this is not a big market to begin with.

Second, some faculty members don't like to use technology so eliminate them from the market.

Third, some faculty members don't have the proper hardware and software platform so eliminate them from the market, too.

Fourth, some faculty members like technology and have the right platform but don't like the teaching approach embodied in the courseware, so eliminate them from the market.

Fifth, some faculty members are perfect candidates but can't afford the software at the offered price, or their students can't because they are already paying big dollars for their textbooks, so eliminate them from the market.

Sixth and seventh, since the market mechanisms aren't perfect, some faculty members simply don't hear about the software and don't buy, or they steal it, so eliminate both of those groups from the market.

As our inquiries proceeded, we did find some cases of valuable, viable courseware but we were surprised by what most of those cases looked like. They often appeared to our sophisticated eyes to be rather small and crude software packages that were not particularly likely to revolutionize anything. Some of the crudeness, of course, came with their viability; in the world of textbooks one knows the texts that have thrived for decades, but in the courseware world, it is the newest prototype that runs on the newest machine that becomes the gauge of quality. However, we came to see that the smallness and crudeness were also the strengths of such courseware. By not being revolutionary, they were more widely and readily acceptable to faculty members who could use them to make marginal improvements in current courses. By being small and crude (i.e., inexpensive to develop) they were also inexpensive to upgrade and to port new to computers and operating systems (for the purposes of brevity, I will henceforth refer to any such improvement as an upgrade).

The failure to upgrade (i.e., to create a new version of the software that can run with a more advanced version of the original machine or operating system, that has more advanced features, or both) sharply limits the lifespan of software, and thus its potential to create a curricular revolution. Courseware must be upgraded every few years or it will die; the underlying hardware and operating systems change, so, after a time, courseware that was not upgraded no longer runs on available machines. If courseware lives the short life of a mayfly, curricular change on a national scale unfolds on a geological time scale: on a national scale it may take five, ten or more years for a course to change. Thus, if courseware is not upgraded regularly, it is unlikely to live long enough to power curricular change.

Unfortunately, the cost of upgrading has proven to be a significant fraction of what it cost to develop the software in the first place, despite hopes that new languages and new authoring systems will change that. Because income from courseware sales tends to be low, there has been an almost iron law killing expensive courseware: if it costs a lot to develop a piece of courseware, upgrades will also be expensive. Yet the income yielded by version 1 is unlikely to pay for an expensive upgrade. Thus large, expensive pieces of courseware rarely reach version 2.

In contrast, courseware that is inexpensive to develop and simple in design (i.e., seemingly crude) is more likely to be affordable to upgrade. Such upgrades can be "financed" by the continuing passion of a developer who has an independent income. Returns from sales (referred to by one of my developer friends as "pizza money") also help.

Non-monetary motives such as passion are important in understanding the abortive life cycle of most courseware. Faculty members and their students usually take a lead role in developing courseware. These people usually have more incentive to develop prototypes or Version 1.0 than they have to develop an upgrade. Version 1.0 is an innovation, perhaps even worthy of notice when promotions are being considered. In contrast, version 3.0 is rarely either intellectually rewarding or worthy of another promotional notch. Universities and their funders (corporations, foundations) are also more attracted by the appeal of starting something novel than they are by the less glamorous, equally expensive labor of bringing it to completion, marketing it, upgrading it, and supporting it.

Note that the Valuable Viable Software Project did not have the resources to study videotape courseware, and we analyzed only one videodisc case study (it was viable). Video is a more viable medium than computer software in that standards for its hardware have been relatively static, so there is no technological reason not to continue using version 1.0 for years. The topics of video-based courses (telecourses) are usually carefully chosen so that there will be wide use, and little dating over five or more years. A second reason for the viability of some video courseware has been money. Video is often used as a vehicle to increase enrollment, a more tangible, income-related motivation for administrators than improving learning. For reasons such as these, a few institutions have found it possible to spend and recoup hundreds of thousands of dollars on developing, using and marketing television- based courses.

At our first meeting, the VVS study group saw the apparent paradox: we knew of few success stories about the kind of software that Levien and others in the decades since had cited as justification for putting computers into universities and colleges, yet those same colleges and universities were filling up with such hardware, and those computers were being heavily used. What was it being used for? and were such uses of educational value?
THE VALUE AND VIABILITY OF WORLDWARE

When the Valuable Viable Software Project team first convened and began to share what we knew about potential case studies, it was immediately obvious that the vast majority were pieces of software originally designed for uses other than undergraduate instruction. Because their origins and chief market were in the real world (including the world of research), I suggested the name worldware.

Worldware includes productivity tools (e.g., word processors, spreadsheets) and other software used in the profession, e.g., molecular modeling tools, computer aided design, software used by professional musicians, software used to manage and search research collections, the databases themselves, communications packages, the internetÉthe list is long and its penetration into undergraduate instruction deep. The VVS book includes several case studies of worldware, including one of the most viable academic software packages ever distributed: the Student Edition of Lotus 1-2-3. Its publisher, Addison- Wesley, estimated in 1992 that this single package of software had been used by 600,000 learners in about 2/3 of all colleges and universities in the U.S., in courses across the curriculum, since it appeared in 1987.

The VVS team decided to make student edition a separate category of software. We defined student editions to be software that is to some extent modeled after worldware but which is marketed explicitly for instructional purposes, and perhaps has been modified or developed for instructional purposes as well. Sometimes the only difference between a piece of worldware and a corresponding student edition is price, but often the student edition is made easier for novices to use (by giving it fewer functions and choices, a more intuitive interface, or added help functions, for example). We discovered that the interaction between student editions (which are sometimes developed by faculty members) and worldware is a rich one. Minitab is statistical software that was originally developed by faculty members as a student version of the kinds of commercial software used by professionals (e.g., SAS, SPSS). Its ease of use and power soon made it popular as professional software, by which time some of the faculty members and graduate students who developed it had started a company to market and improve it: Minitab, Inc. Today, closing the circle, there is a student edition of Minitab, developed by Addison-Wesley.
WHY ARE WORLDWARE AND STUDENT EDITIONS SO VIABLE?

The VVS team had several reasons for its conclusion that worldware and student editions are usually more viable than distributable courseware:

1. Whether worldware is developed for (well-funded) faculty research or (lucrative) business applications, multiple producers are probably racing with one another to produce faster, more powerful, less expensive packages that run on multiple platforms. New versions will usually run files from older versions, and even from competitor's packages. Faculty members can be confident that, if they invest time in rethinking portions of a course to take advantage of worldware, they can use the new course materials for as long as they choose.
2. Worldware is often multi-purpose, which makes a bigger market. Because most worldware takes the form of tools or resources, rather than being shaped for instruction, it is likely to be more flexible in application than courseware. The same worldware can be used for different purposes by faculty members with entirely different instructional approaches to the same course and for many different courses as well.
3. Students come in already wanting it. If people in the real world are using new tools to think in new ways about their problems, and even to tackle new problems, they will soon expect college graduates to already have learned the rudiments of those new ways of thinking. The undergraduates coming into the classroom will be expecting to learn them, too.
4. Faculty members already know about it, too. Because of the software's popularity in the real world, faculty members may already be familiar with it, and perhaps even use it in their own research or consulting. Use of the software in the real world lends credibility to its vendors' claims for its reliability, ease of use, and so on. Because it is already used in a wide market, technical support is likely to be superior to that for courseware, too. Thus there is less marketing and training to be done than with courseware. That worldware is in common use is no surprise, but many people, eyes still fixed on the dream articulated by Levien and others, have decried it as a distraction of little educational value. "Our machines are in use, but merely for word processing." But as many faculty members have already concluded, worldware can be of substantial value in making major improvements in courses.

THE VALUE OF WORLDWARE AND STUDENT EDITIONS

Consider some of the following courses transformed by faculty members because they and/or their students could use worldware:

* Statistics courses, no longer limited to obsolete paper and pencil techniques, can now teacher more powerful statistical procedures. Faculty members can now put students to work on real world problems, in the service of real world clients, with all the engagement and student energy that can elicit (see example described in Lewis, 1989).
* Foreign language courses can now more readily teach students how language is spoken in a foreign culture, and about the culture itself, thanks to the availability of video programs from that country, available by satellite, cable, or services that obtain rights to such video and package it with instructional support (e.g., PICS at the University of Iowa).
* Universities could now, if they desired, teach undergraduates new skills for personal use after graduation, e.g., designing rooms for their homes and offices (now that they can use computer-aided design), composing and playing music using computers, creation of art using computer graphics. It is crucial to repeat that the software does not teach students skill in such design. But the software does create new opportunities for faculty members to teach such skills to students whose talents would not have allowed it when only traditional design tools were available.

Once you look, it is easy to find classrooms in US colleges and universities where the content and objectives of today's curriculum have become dependent upon the use of worldware. Is that the revolution we have been hoping for in learning? The following story, for me, suggests what the revolution may look like.

Six years ago, Raymond J. Lewis and I were looking for faculty members who had had at least two years of teaching in an environment where students had unfettered access to personal computing. In 1987, such faculty members were still quite unusual, but it seemed likely that within a few years their numbers would explode. We wondered what these pioneers might already have learned about the benefits and difficulties of teaching in a tool-rich, information-rich environment.

One place we visited was Reed College in Portland, Oregon, where the current seniors had had four years of easy access to Macintosh computers. In one day, I talked to faculty members from eight departments, asking what they liked about teaching in this environment.

Surprisingly, there was one thing that all of them had noticed. As two of them put it, "I'm no longer embarrassed to ask the student to do it over again." Because computer- based documents and projects are mechanically easier to revise, their students pressed to get a second chance to improve their work and their grade. Gradually the texture of the curriculum in each course was changing: toward fewer projects, with each project being developed in stages -- plan, draft, conversation, another draft, final version. Each stage of work was marked by rethinking, and learning. ,

That day Ray and I also talked asked some seniors if they thought their education -- what they had learned -- had been influenced by their use of computers. One of them replied that he'd learned that it's not one's first draft or thought that matters, but the last one. He had learned that lesson over a series of courses, he said. Similarly, several faculty members and the director of the writing program each suggested that the most tangible impact of Reed's four years of making computers available to this graduating class would be found in the quality and coherence of their bachelor's theses.

The day at Reed had a bit of a surprise ending for us. When Ray and I sat down with several of the Colleges educational and technology leaders, they were all surprised to hear about what we had heard. This phenomenon, which we called "Doing It Again, Thoughtfully (DIATing)", had been an invisible, ecological change, not directed centrally, but powered by independent actions by many faculty members and students across the college.

This story suggests at least three lessons about worldware and the educational revolution we hope to see from technology:

* Worldware can enable important changes in curriculum, even though it has no curricular content,
* What matters most are uses of technology that influence the student's whole course of study, and worldware is flexible enough and relevant enough to make such change possible, and
* If such curricular improvements come from independent choices made by faculty members and students, the cumulative effect can be significant and yet remain invisible to university administrators and to national experts on education.

AND THE WORLD WILL BEAT A PATH TO YOUR DOOR (?)

At FIPSE in the early 1980s we had already begun to glimpse the promise of such software. Our glimpse taught us that worldware by itself was unlikely to cause an educational revolution either.

In 1982, before Lotus 1-2-3 had even been introduced, Professor Harold Cochrane of Colorado State University wrote a proposal about computer spreadsheets to FIPSE. Cochrane had realized that spreadsheets (at that time thought of strictly as tools for accounting) were multi- purpose mathematics tools. With the aid of the FIPSE grant, he developed an economics course in which students used spreadsheets to simulate economic phenomena. Sometimes the students themselves would play roles in the simulation, using the computer to calculate the macroeconomic consequences of their microeconomic actions.

Cochrane's ideas had exciting possibilities for helping students to understand transport phenomena and other ideas in fields from chemistry to political science, and FIPSE staff encouraged him to make presentations not only within economics but also in venues such as the annual meeting of the American Association of Higher Education. Then we waited for the idea to spread like wildfire to other faculty members in economics and other disciplines. And waited.

There were at least three barriers blocking the rapid spread of Cochrane's model for using what today we would call worldware.

The first, of course, was the lack of machines. Cochrane's idea required computer labs in which classes could be held, enabling students sometimes to work in small groups around computers and sometimes to convene as a class; such labs were few and far between in the mid '80s and for years thereafter.

A second barrier was the much-complained-of lack of incentive for professors to take risks in improving teaching.

The third and equally important barrier is learning by faculty. It is not obvious how to use spreadsheets to teach ideas in chemistry, for example, nor is it obvious beforehand what might go wrong. Worldware is a particularly problematic teaching tool in this way because the spreadsheet does not come with a manual of teaching ideas.

None of these barriers could permanently prevent the idea from spreading. Physics professors, for example, soon began drafting manuscripts for textbooks teaching students to use spreadsheets to solve physics problems. But publishers would not publish them. It took even more years to convince publishers of the market. It was a chicken and egg problem, the frustrated faculty felt. How could there be a market without textbooks, and how could there be textbooks without a market? Finally the textbooks began to appear and now there are a number of them. But the process was painfully slow.

Sometimes the government can take action to speed curricular change based on worldware. The National Science Foundation several years ago saw that new software such as Mathematica would make traditional calculus courses obsolete because the software could do automatically operations that students were painfully learning (over many months) to do with paper and pencil techniques that they would never need again. So beginning in the 1988 fiscal year the Foundation began a national effort to transform the teaching of calculus, supporting the development of experimental courses, eight new textbooks, workshops for faculty members, and a variety of other efforts. The program was successful enough that when the textbook Calculus, developed with NSF funding by faculty members at eight institutions and tested at 50 institutions across the United States, was published, its initial printing of 18,000 copies sold out in a few months.
PROFESSORS NEED MORE THAN JUST "HOW-TO": THE CASE OF ENFI

Thus far we've assumed that the challenge of using worldware is limited to issues such as designing new syllabi, homework assignments, and tests. Another piece of history shows that professors have even more to learn if "revolutionizing" their courses with worldware is to become less risky and time-consuming.

In the late 1980s a new technology-enabled teaching technique was developed in composition. Students and professors write to another over a computer network, rather than speaking. The class dialogue continually scrolls up the screen. The advantages of this technique (called Educational Networking For Interaction, or ENFI) are many. A student at Carnegie Mellon observed that, after a time, he forgot he was writing, which is one of the objectives of ENFI: to help students internalize the idea that writing is not just an academic exercise.

Professor Trent Batson, developer of the ENFI family of teaching techniques, convened a group of professors from various colleges using the technique for a continuing series of discussions on its implications. In those discussions, they made a surprising discovery.

After the professors got comfortable with one another, one had the courage to tell an embarrassing story. Early in his course, students had erupted in (computer-based) profanity and obscenity. The faculty member, constrained to "speak" only on the computer screen, could not regain control of the class and had walked out. It had taken several meetings of the class for the mortified faculty member to regain to control. To his surprise, several other professors in the room admitted that the same thing had happened to them.

For each faculty member victimized by this phenomenon, it had seemed the result of some obscure personal failure. Only when the professors talked with each other, discovering that this was not a unique occurrence but something that had happened to many of them, were they able to discover why it had happened. In the traditional classroom, the faculty member can hardly avoid taking a dominating position. Standing while students sit, or even sitting with them in a circle, they force the center of attention back to the faculty member, and attempt to force even a very learner-centered teacher to tell them what to do. Things are rather different in an ENFI lab setting. The faculty member may, like the students in the circle, be virtually hidden behind the screen of his or her computer. The faculty member's utterances ("Let's get back on the subject shall we?") are just one more line of text swiftly scrolling up and off the screen. The students, encouraged to write to one another, sense freedom and explode. The good news is that, once these professors had discovered that the universality and roots of the phenomenon, they were also able to share ideas for how to harness this energy in the service of the composition course. And so they progressed.

This story may seem an isolated incident, but I believe such turbulence and disruption are virtually inevitable if the curricular "revolution" is more than trivial. Teaching a particular course is a craft. Crafts usually progress through the cumulative learning of artisans. When underlying conditions change, even a little, the craft may be set back. That's the situation faced by faculty members using technology. Adopting a new technology and trying to change the curriculum may result in worse teaching, not better, because they are working in an unfamiliar medium with hidden pitfalls. Colleagues who are trying to change the same course in similar ways can provide a light to reveal some of those pitfalls, as well as subtle opportunities (such as turning the energy behind the profanity in ENFI classrooms to instructional use). If worldware is to be exploited, we also need new textbooks (or the equivalent). The canonical version of a course (and thus its materials) cannot change until enough professors are visibly ready to adopt new kinds of materials -- enough so that publishers develop and deliver those materials, make money doing so, and begin to deepen and widen their investments. In the calculus initiative described above, note the NSF's effort to get publishers involved, racing in competition with one another to bring high quality textbooks to market, and to market them.

In sum, if a worldware-enabled curricular revolution is to happen, large numbers of faculty need to collaborate in the pre-textbook phase of change. That does not usually occur because faculty do not usually have the means or incentive to work together in that way. With that despairing thought, we will leave the worldware revolution for the moment, returning to it in the next chapter, "Higher Education, 1998."
NEW PATHWAYS TO A DEGREE

My second misconception in the late 1970s, as I remarked at the start of this chapter, was that goals of quality and accessibility were necessarily in conflict. It did seem obvious at the time.
THE CONFLICT BETWEEN QUALITY AND ACCESS, COMPUTING AND VIDEO

In the 1970s and early '80s, computer courseware was ordinarily used to enrich quality, often in a rather spectacular way but on machines available to only a small minority of students. Meanwhile video courseware was ordinarily used to extend access to students off-campus. Further, the advocates of computing and video (telecourses) disagreed on more than just which educational need was more critique: improved quality or accessibility. Users of computing tended to value active learning and individualization, and consequently deplored television as a passive, mass medium. Users of video tended to value video's ability to add a visual dimension to learning and to offer that learning to a mass audience, while consequently deplored computing's crude graphics and elitist nature (expensive machines for the few).

No one was surprised at the conflict between advocates of quality and accessibility. Academic resources have been usable only in one place at one time by a few people. Therefore universities have always faced the cruel choice: concentrate them for the use of the few, or allow the many access, thus diluting (and perhaps degrading) irreplaceable resources. The choice seemed most simply expressed in the student-faculty ratio: some wanted to raise it (e.g., through video of the best teachers) and others to lower it.

Although few faculty think of it this way, technology has periodically been used to redefine this conflict between quality and accessibility. What is the student-faculty ratio of a book (once seen as a dehumanizing, dangerous technology)? Books increase the ratio of students to professors by spreading the words of each expert to more students. Books increase the ratio of professors to students by giving each student access to far more expertise. Thus the technology of the book can improve both the quality and accessibility of educational resources.

Until recently, professors, books, and laboratories could only be in one place at a time. Since they are rare, expensive, and fragile it has made sense to collect them in particular places (i.e., campuses), wall them in for protection, and limit admission to their use. This need for cloistering of academic resources dictates other features of academic life such as scheduled courses, semesters, tenured professors working within the campus' boundaries, and so on.

The Annenberg/CPB Project's mission is to use technology to redefine the seemingly inevitable quality-accessibility tradeoff. The Project was commissioned and funded by Ambassador Walter Annenberg because he believed that a high quality of education could be made available at reasonable prices by the mass of institutions if they could use technology-based materials of extremely high quality. His grant to the Corporation for Public Broadcasting made it possible to develop courses that each cost $2-7 million dollars. The resulting instructional materials were not "talking head" video. "French in Action" and "Destinos" are foreign language courses organized around engaging stories shot in countries that speak French and Spanish, respectively. Part of their power comes from reminding students every minute of why they are learning the language: so that one day they can travel to these countries, and understand what they see and hear there. "Mechanical Universe and Beyond" combines extensive computer graphics and location photography that help students visualize physical phenomena and the workings of the calculus with a healthy dose of the history of the field. "War and Peace in the Nuclear Age" is based on a wealth of historical footage and interviews with people who made the history of our use of nuclear weapons.

The development of such magnificent course materials is highly dependent on the generosity of governments or private donors who are willing to spend millions of dollars, however. It is not an accident that the Annenberg/CPB courses do not add up to a bachelor's degree program. Not even Ambassador Annenberg's generosity was sufficient to develop, market and support courses that would each be widely used while also providing a coherent path to a four year degree. For that matter, the "same" upper division courses taught at different U.S. institutions are so diverse, and their content so rapidly changing, that even if the money had been given to develop television- and print-based courses at this level, their degree of use would have been disastrously small and short-lived.
"BOTTOM UP "VERSUS "TOP DOWN"?

Courses can be changed by the will of individual faculty, "bottom up" (if you are a student of organizations who doesn't realize that U.S. institutions of higher education can be inverted pyramids, with the faculty on top and the president on the bottom). Virtually all of U.S. investments in curricular software, video or computer, have focused on assignments or, at the most, full courses, often at the lower division, at levels under the control of individual faculty.

In the early 1980s FIPSE funded one of the first efforts to take advantage of technology to rethink a course sequence leading toward a degree. Marvin Marcus and his colleagues at the University of California, Santa Barbara, developed courses in applied mathematics that included new content made possible by microcomputer laboratories and a field component that gave students experience in using their new math skills on real world problems off-campus. Later in the 1980s, the MUPPET program in physics at the University of Maryland began the FIPSE-funded process of considering what a physics degree program should look like, if one could assume that students could use computers to create simulations and do calculations. The Maryland program, significantly, relies mainly on student use of computer languages (worldware) to learn to create their own tools and simulations. In the meantime, the use of worldware began to aid subtle curricular change in many fields, from engineering to political science (use of databases, word processors). Their curricular software consists of program elements (e.g., elements of simulations) that students can tailor and expand upon.

Despite a few such pioneers, as the 1980s ended, there were few departmental and institutional leaders rethinking degree programs and their supporting services (admissions, libraries, financial aid, etc.) in light of the capabilities of new technologies.
SINGLE VERSUS MULTIPLE TECHNOLOGIES

Campuses are collections of multiple technologies, and we think it proper, not unusual, if a course involves an auditorium and a professor and an overhead projector and use of a library and use of a laboratory. Yet the image of new technologies has implied reliance on single courses (e.g., terms such as television-based course or computer- aided instruction implying that no faculty or print were required). Until the late 1980s, electronic technologies had generally been expensive and arcane enough that the masters of the technology controlled the curriculum. Some institutions offered several different sets of courses for learners off-campus: a television-based curriculum, an audio conferencing curriculum, a correspondence (print) curriculum, and a modem-based (electronic mail) curriculum, for example.

As the 1980s drew to a close, however, a few conference presentations here and there revealed that some institutions were experimenting with courses using several different technologies to give learners off-campus a rich and well-structured experience, e.g., a combination of live or taped video for lectures, audio conferencing for small group discussion, and electronic mail for discussion and transfer of homework.
"BIT BY BIT, PUTTING IT TOGETHER"

My colleagues and I at the Annenberg/CPB Project (especially Dr. Mara Mayor, Dr. Lin Foa, Dr. Scott Roberts, and Dr. Michael Strait) saw these trends toward use of multiple technologies and toward thinking in terms of degree programs rather than just single courses and decided to lend a hand.

Thus, in 1990, we created a funding program called "New Pathways to a Degree: Using Technologies to Open the College." Working adults and many others off-campus needed access to an intellectually challenging education," our guidelines asserted, and those learners required support in four key areas:

1. Access to courses and academic services that are available at convenient times and places;
2. Resources at least comparable to what they would find if they were full-time students on campus -- including stimulating teachers, collections of rich primary and secondary source materials, and laboratory experiences;
3. Opportunities to discuss ideas and exchange homework and other materials with professors and students;
4. A coherent and substantial array of courses that enables them to move efficiently toward the baccalaureate.

We had several reasons for this unique focus on technology-enabled course sequences leading to a degree:

1. Our reading of the research literature indicated that single courses rarely, by themselves, have intrinsically valuable, lasting impacts on the mass of their students. The things the mass of students learn and later use almost always come from a sequence of courses, often reinforced by key elements of the college climate and extracurricular experiences. Thus we were interested in intellectually coherent sets of courses.
2. By the same token we knew that it was futile to offer courses across barriers of time and space if key student services (acquisition of books, library services, counseling, etc.) were not available in the same way. Institutional investment in accessible services was more likely if a whole degree program rested in the balance.
3. Acquisition of hardware, and skill in its use, is key for a program like this. Students, professors and administration were all more likely to undertake such efforts and such risks if they could be amortized over a whole series of courses. We felt students were even more likely to enroll if they knew that they could complete all, or most, of the requirements for a degree.

The program offered grants of $150,000 for two year projects by single colleges or universities, and $300,000 for groups of institutions. Applicants were challenged to provide substantial cost-sharing. As program manager, I had to guess how many proposals we might receive for the $1.5 million we were offering. Experienced in funding programs for innovation, I made two predictions: a) we would receive about 70 proposals, and b) most of them would not be much good, since this was a new field.

It is significant how wrong I was. We received 243 proposals, representing well over 10% of all the colleges and universities in our country (many came from groups of institutions). And perhaps 90% of the applicants appeared quite capable of using the money to create programs that would meet the four criteria listed above. Another interesting point: almost all of them proposed to use the new technology of fax machines, despite the fact that not one foundation or government program had any special funding program to integrate the fax machine into academic practice. Obviously there was a great deal of change in the country that was not being spotlighted by grant programs. In our earlier discussion of Reed College, we saw how educational progress could change a college while remaining invisible to that community. The New Pathways solicitation drew such an amazing number of good proposals because faculty members and administrators all across the United States had been invisibly gaining experience in using a variety of technologies and in serving adult learners off-campus. When the Annenberg/CPB Project had the good luck to offer a program at the right moment, they flocked to it with well-grounded proposals.

The Project was able to fund seven projects. Three brief observations should give you a glimpse of their common points and their variety:

* The University of Maine at Augusta leads a statewide effort to offer two year degrees using video, computers and telecommunications. Students from throughout this highly rural state can travel to nearby high schools to watch live classes by video, and talk with the professor and other students by audio link. These students have access to unprecedented library riches; like other students in the state today, they can consult a unified on-line library system. When their research turns up titles they need to read, the University will mail them the books, no matter what they live in the state. In contrast, the Rochester Institute of Technology helps its faculty tape their lectures before the start of the term. Students rent or buy the tapes, and then study them at convenient times and places; audioconferencing and audiographic conferencing are used for real-time discussion of the lecture materials. Both Maine and RIT depend heavily on worldware, especially hardware (computers, video systems, library management systems, etc.)
* At Indiana University - Purdue University at Indianapolis, one problem was how to deliver a chemistry lab to their off-campus sites. Dean Erv Boschmann and Susan Frantsi developed a lab sequence that combined micro quantities of chemicals, videodisc simulations (one of the few examples where one of these projects chose to use faculty-developed computer courseware from another institution), and experiments done with household materials. His colleagues were so impressed with the power, cost-effectiveness, and safety of the lab that it is being used for many students on-campus as well.
* All of the programs make at least some use of asynchronous electronic communication among professors, graduate assistants, and students: typically both electronic mail and fax. As I will discuss in the second of these two essays, research and anecdotes hint that this medium provides instructional advantages not available using traditional exchange and conversation mechanisms on- campus.

It is also intriguing that many of these seven projects have begun to move toward more integrated support of technology, combining video, computing and telecommunications in one office or division, often joined with other functions such as libraries or instructional design.
CONCLUSIONS

If I were simply to correct my original misconceptions, so that they became new conceptions that made sense for the 1990s, they might look something like this:

1. Although distributable courseware has a role to play, worldware and student editions are the key today to making substantial improvements in teaching and learning. Worldware is inexpensive enough and potentially revolutionary enough to figure in the reshaping of large numbers of courses at large numbers of institutions. It is too early to celebrate, however. Worldware and student editions do not come with instructions on how they can be used to transform a course. Somehow that missing element must be supplied.
2. Computer- and video- courseware can be used in combination to enable academic programs that both extend accessibility and enrich quality. They are not in conflict. They have complementary strengths.
3. Courses and academic programs continue to rest on the use of multiple technologies, some electronic and some not. Course design needs to exploit the best such ingredients, not maximize the usefulness of one at the expense of the outcomes for, or pocketbook of, the student.
4. The creativity of individual faculty improving their courses for their students is the single most powerful determinant of the use of technology, but it isn't enough. The university and its friends must also attend to larger issues: the content and structure of degree programs, for example, and the organizational structures and services that support them.

Unfortunately, those four revised
---- Original electric version breaks off here. ----
---- Remainder supplied from hard copy. ----
statements are only fragments of a vision. Given what we have learned since the 1970s, how should higher education use technology to educate more students, better, for a supportable budget?