The Rise of the Virtual University Free

As we rush to put courses on the Web, it is helpful to remember that existing college courses aren’t so wonderful right now. Try to remember when you were in school. How much do you remember of what you learned? We all memorized and then forgot what we memorized days later. What we do remember from college is typically stuff we have used since. If we don’t use it, we lose it, and most of what we learned wasn’t all that usable in the first place. There are two issues here. One is how to teach in a way that makes things memorable, and the second is how to decide what to teach in the first place.

One reason we don’t remember is that it is actually very difficult to simply listen when someone talks. In the best case, we think about what the lecturer has just said at the same time they continue to talk and thus we miss a lot of what they say to us. To connect what they say to what we think requires us to stop listening and think a bit. In the worst case, we are thinking about something else entirely. Professors know this, which is one reason why there are tests. Simply put, lecturing is antithetical to learning. It is very hard to pay attention to an hour-long barrage of information. We can be entertained by an hour presentation if it is indeed entertaining, but we simply don’t learn very well what was said to us. Retention is based on use.

People learn by doing. This is a well-known idea that has been around for a long time. In fact, John Dewey (1916) lamented that, even though everyone knew that people learn by doing and cannot “learn by pouring in,” there seemed to be no way to change the schools. Well, now there is. Learning by doing needs a medium and computers can be that medium. Existing college courses, when they allow for doing can succeed well. But a great many courses, especially introductory courses and service courses, have little or no doing in them at all. This can, and will, change when the Virtual University ascends.

Current college courses fail not only in their means of delivery, but also in what they are trying to deliver. This is true for a variety of reasons, the two main ones being the idea of a curriculum and the concept of service courses. Colleges have the sense that they know what students should learn, so they create curricula that require students take a course in X or fill the Y distribution requirement with a number of possible courses. So students who are interested in learning to do Y find a set of hoops to be jumped through in order to do Y, including a variety of prerequisites.

The problem is that every time students takes a required course, they find themselves faced with a serious motivation problem. By this I don’t mean that they aren’t interested in the course and can’t get themselves to care about it, although this may well be true. What I am saying here is that if you don’t know why you need to know something, it is actually quite difficult to learn it, and what you learn won’t stay in memory for long. If we don’t use something, or at least see how we might use it, it is very difficult to retain it. Thus, a course in calculus may well be useful for an economist, but since the course will likely have very little to do with economics, and the calculus that will be used by the economist will not be used until some later time, the student will likely have trouble caring about or retaining what he learned. If he never uses what he learns, he will forget it entirely.

How long should a college course be? In the United States, the current answer is about forty hours of seat time spanning a twelve-week period. Is there a reason for this? Does some great educational point underlie it? No. It simply has always been done this way. How long should a course actually be? As long as it takes to learn to do what the course is teaching you to do. This is so obvious it seems almost absurd to mention it. And it would be absurd if it weren’t for the fact that this ideal is violated in nearly every college course. The one big exception here is the Ph.D. thesis, which takes as long as it takes. By and large, graduate education in the U.S. is much closer to a learn-by-doing philosophy.

Courses that take as long as they take are really not possible in current university environments. Each student can’t get his own length course; professors can’t be available for as long as this takes. There had to be some standardization on time. Not any more. The Web changes all that: take a course when you want and learn as much as you want until you can demonstrate that you can do what the course is trying to teach you to do. This makes sense. Naturally, this is not what current Webbased courses are doing, for the most part. The current courses look like copies of the old ones. Just as early films were just filmed plays, early Web courses are just regular courses on the Web. This will change, and when it does the structure of the university system will have to change with it. You won’t need so many course credits to graduate because the concept of “credits” will have become meaningless. Credit just means number of courses taken. What is needed is a new concept, based upon performance. Graduates should have accomplished certain things, not necessarily have sat through certain courses.

For all its promise, education via the computer has been, by and large, a disappointment, overhyped and under-realized. Computer-based training has meant putting a book on a computer, allowing the student to press a button to get the next page and take a quiz at the end. Often, instructional software, purporting to be teaching valuable facts to children, has kept parents happy because kids use it; it looks educational and it keeps kids happy because it’s a video game disguised as education, all of which means parents don’t disapprove of their children using it. What does it teach? Often, very little.

Can we do better? You bet we can. The air flight simulator is a very good piece of educational software; there is no better way to learn to fly (that isn’t dangerous). Learning by doing is a practical reality given good simulations. The problem is both to build those simulations and to reinvent a curriculum based upon this new technology and the idea of learning by doing.

The rise of the Virtual University (VU) allows us to reconsider education. It is reasonable to assume that education involves learning and, by and large, we are all under the impression that school is a place where learning goes on, or at least it is a place where learning is supposed to go on. Business is so dissatisfied with the state of education today that we find corporation after corporation creating their own “universities” where employees may attempt to relearn what they failed to learn in school or where they can learn what they might need to know on their jobs. Even so, those businesses that have recognized that school didn’t work out for their employees are nevertheless of the belief that school is the right model for learning. Curiously, business training tends to be a copy of existing school systems.

We have the opportunity to create some massive changes in what it means to be involved in obtaining an education. To do this, we must change the model of school completely. Many professors in today’s universities are not motivated to provide high quality teaching. They know students will not act like consumers, despite the fact that they are paying the bills. Rather, because students need the certification and recommendations universities provide, professors are in a power position, not in a service provider’s position.

Professors understand that they can dominate students and create various hoops for students to jump through in order to get a good grade, but that they don’t really have to worry whether anyone has learned anything. In this model it is all too easy to just lecture and test and forget about real education.

This is all okay with students, as it turns out. There is an implicit gentlemen’s agreement about school. Teachers make demands, students satisfy those demands, and those who play the game by the rules win. “You give me the grade, I’ll get the degree, and I’m out of here.”

To understand why universities are the way they are, we must understand the following things: why professors teach, what professors teach, what students expect, how employers view the graduates, and the role of certification. We will now tackle each of these issues.

There is a certain naiveté on the part of students in universities about why the professors who teach them are there. They assume teachers teach because it’s their job and that the model that held in high school of the professional teacher applies to the university as well. Rather, professors at the top universities teach because they feel they have to, or ought to, rarely because they see teaching as fundamental to their life’s work. At my university, professors who don’t get research grants or contribute to the university in other ways are “punished” by teaching. The worst professors—that is, the ones who care the least or who have tenure but gave up research a long time ago—typically teach the most courses. A top-notch professor, one who is world famous and brings in lots of research dollars, may teach as little as one course every two years. On the other hand, colleagues who have none of these attributes may teach as many as four courses a quarter.

It would be nice to imagine that this is a good system for students, hoping that the nonresearchers happen to be good teachers, but in fact the opposite is the case. Typically they are burnt out researchers, resentful of how they fared in the system and with little understanding of how their field has changed in the last 30 years.

The best professors in the U.S. may or may not be the best teachers. This is actually a complicated idea because the issue of what defines “best” is subject to question and what defines good teaching is a very open issue. In the competitive world of American universities, “best” has a clear meaning. Universities vie for the services of professors who have the biggest reputations. Top professors get great deals, as they are sought-after by the top schools. These deals include, of course, higher salaries, but as universities can only go so high, other issues matter as well. One of the biggest is teaching load. As a result, the best professors have teaching loads of nearly zero and sometimes of literally zero. Clearly, in such an environment teaching is not valued, despite what these same universities say to their prospective students.

Nevertheless, these same professors who happily avoid teaching are often the best teachers. The reason why this is so is not obvious to the prospective buyers of these services. Typically, when students attempt to decide between Harvard and Amherst, for example, they say in Amherst’s defense that the professors there are professional teachers and care more about the students and pay more attention to teaching. Like any generalization, this one can be dead wrong, but on the whole it is true. The real question is: of the professors at each institution, who knows more? On the face of it, this seems like a silly question. A course in the classics is a course in the classics. The best teacher would teach the best course, and the knowledge of any teacher is likely to be the same.

While this may be true for the classics, it is not at all true for a field like Artificial Intelligence (AI). Artificial Intelligence as a field is being invented today. The people who are inventing it reside at big and well-funded research labs that allow them to build their toys and experiment with all kinds of hardware. They don’t do this at Amherst. If you want to learn about AI, you can take a course about it at Amherst, but the teacher of that course will be someone who read some books about it rather than someone who is doing it. Actually, Amherst students can sign up for AI courses at the nearby University of Massachusetts which, despite not being a university considered to be of the quality of Amherst, is far better equipped to teach them.

Of course, you could go to MIT (which has a very good AI lab) and take a course from the best professors in AI. But these are the very same people who don’t teach all that much, who don’t value teaching, and who, when they do teach, have classes with hundreds of students in them. The best professors in AI may or may not be good teachers, but they are not professors because they want to be teachers, they are professors because they want to be researchers. They want to build robots or explore how the mind makes generalizations or figure out how to get a computer to be world chess champion. Teaching is one of the last things on their minds.

So, the short answer to the question that heads this section is that in the best schools they teach because they have to, not because they want to, and in the good teaching schools there is a very good chance that they teach what they don’t understand all that well. Now, this is not true in, say, English literature. In subjects where research doesn’t play a big role and where labs are not costly and where professors don’t need graduate students to help with their work, good teachers can be found easily. In fact, those are precisely the fields where there is a glut of teachers. No matter how much professors say they like teaching, and many of them do like it a lot, most recognize that teaching is not what they were hired for, nor is it what they really do for a living.

One of the major problems with today’s schools is the curriculum. When you tell a professor he is to teach a class, you might assume this means he will teach whatever any other professor might teach in that class. After all, one high school history course is like another, so one might assume that this is true of college as well. But curricula in college are professordependent. This wouldn’t be such a problem if it only meant there was slight variance in how a given course was taught from year to year and professor to professor. Unfortunately, the issue is bigger than that.

Professors teach what they know. There is no standard set of things to be taught in anything but the most introductory of courses. So, Introduction to Psychology is pretty similar in every university, as is Freshman Calculus. But, any advanced course is subject to the professor’s unique view on his field. This is fine because one goes to university to meet interesting faculty and to learn what their view of the field is. Or one ought to. This is what universities have to offer; an opportunity to engage a world expert on his own turf to discuss ideas he created or is deeply involved with and for just a few weeks to pretend that you are a world class economist or sociologist dealing with issues just as professionals in those fields do. This is the ideal. The reality is something else again.

One of the problems with this view is that students by and large don’t share it. Most college students go to class expecting to learn the facts. They want to know how economics or sociology works. When I teach a class on how the mind works, students want to know how it works and I should please tell them. The difficulty with this view is that most professors don’t actually know the answers to the questions students pose. Economics professors don’t know how the economy works and sociologists don’t know how society works and I don’t know how the mind works. What we all do have are deeply held beliefs about these subjects.

Students have no idea that this is what they are getting into. They just want to know what is true. They don’t want to hear one man’s viewpoint. But that is what they get every time (unless the professor is a woman). For this reason, one university is quite different than another, and every course in AI is different at every school. This is the fun part of teaching. Professors like talking about their own work and their own ideas. They love talking about the research they are doing. The question is: Is this what students came to learn? By and large, I think, it is not.

Students expect that the curriculum set forth for them by the faculty is meant to help them get where they want to go after school. This simply isn’t true. In computer science, for example, the skills that will get students jobs include various programming skills that are used in industry. One might think that computer science departments around the country would make sure that all these employable skills are taught in their curriculum; indeed, one would expect them to be the center of the curriculum. Sorry. Most computer science professors are not familiar with the commercial packages that are in use on a daily basis in industry and even if they happen to know them, they consider them to be of little intellectual interest. So, a computer science student will learn the mathematics involved in making calculations about what is computable, they will learn the theory of designing programming languages, but they will not learn much of what they will ever use in the real world. Computer scientists want their field to be a science and they want students to attempt to practice that science, despite the fact that the students are there because they want jobs in industry.

There is tremendous dissatisfaction in corporate America with the products of many universities. Sometimes this dissatisfaction is right on and other times it is entirely confused. The dissatisfaction is real enough, but the blame is often oddly placed. What would employers like students to know that they don’t know?

Corporations across America worry about students knowing basic business concepts (like accounting), knowing about how to work in teams, knowing how to write well, make oral presentations, and generally knowing how and why businesses work. But, where would students learn all this? Even a major in business might not learn all these things, and most universities discourage undergraduate majors in business. So, a student interested in business is likely to major in economics, where he learns about macro and micro economic systems and learns next to nothing of what I have listed above.

Of course, I am not recommending that a college education ought to be proper training in business. The problem is really with the conception of a liberal education and the monopoly on education that is held by those who have that conception. Students think they should go to college to get a job and colleges think students are there for some other reason entirely. A compromise might be nice. Colleges do have some obligation to raise the consciousness of students beyond their initial aspirations. On the other hand, they also have the obligation to respect the practical exigencies that are extant in today’s world.

Political science majors presumably want to work in politics and usually do not want to work on the theory of political systems. Psychology majors presumably are interested in the mind and might want to work in health-related fields and are not likely to become experimental psychologists. However, professors often share the idea that they are really training their students to become academics like themselves and that their job is to cater to the one or two students who show promise in that regard. All other students—those who will become practitioners in these fields—are given short shrift and not taken seriously by the curriculum committee. Individual professors can and do work around the system they have set up, but by and large the system does not enable or even care about future student employment.

Universities will never grow out of their certification mission. Too much depends upon it. It is hard to imagine that as many people would go to college as do now, if no one really cared about whether you had been to college. No one would fight to go to Harvard if going to Harvard didn’t matter. But what matters about it? Not the education. No one asks if you learned a lot, they just assume you are smart because you went there. It is time to rethink this.

We won’t get rid of certification, but perhaps we can contemplate new kinds of certification. Students should be certified as having accomplished something or as being able to do something. Like Boy Scout merit badges or karate black belts or truck driver’s licenses, the proof should be in the pudding. Students should show their stuff; they should be able to do something and the attestation to the doing should be the certification.

Such changes are unlikely to occur in current universities. It is the rare faculty member who will willingly stop teaching the same old course he or she has taught for 30 years and design a new one that will be more work to teach because it requires more individual effort. This will not happen unless the venue and the circumstances of education change radically.

Here, then, is why we can begin to have some hopes for the Virtual University. It is not the case that these changes can occur only at VU. They could occur anywhere, but they won’t. The inertia is too great at State U. VU has such promise because VU is virgin territory. There is no entrenched establishment that will block change. John Dewey would have been ecstatic.

In order for real educational change to take place, we must first abandon the basic model of education that holds that education means the accumulation of facts. This is because, in a deep sense, education isn’t about knowledge, and it isn’t about getting students to know what happened. It is about getting them to feel what happened. This is, of course, not so easy to do.

I recently had a conversation with a divorce attorney who was advocating an elementary school curriculum that would help one choose a mate. He called it a “preventive maintenance” curriculum meant to prevent divorce by exposing children to the issues of marriage. If you want children to understand how to avoid bad marriages, it is not unreasonable to force them to live with somebody for a while. No one would think to do this because of the practical realities involved, but make a boy and a girl live together for a week and they will learn a great deal about marriage.

This is an example of natural learning. We all learn about relationships by experiencing them. Could we teach such things in a classroom? It seems obvious that it would be quite valuable for people to study human relationships and how they work best, but unfortunately school isn’t really about that. But why not? And what would learning look like and what would school be like if ideas like the attorney’s were implemented?

What school is and isn’t about can be discussed endlessly. I will get into more detail on this subject later on, but for now, let me just make the obvious point that there was no divine intervention when it came to deciding upon the school curriculum. No deity proclaimed that major subjects of study were to be mathematics, science, history, language and literature. Yet, whenever we attempt to modify this curriculum, the defenders of the faith are loud and numerous.

When we begin to think about learning, we must immediately ask about knowledge. It is very difficult to think about education without thinking about the knowledge we want to impart to students. We live in a world in which knowledge reigns supreme. In the popular culture, games like Trivial Pursuit capture the country’s attention; television focuses on Jeopardy and other “knowledge games” that test who knows what. And, far more importantly, school focuses on the same “trivial” knowledge. Schools are driven by tests and these tests focus on fill-in-the blank and multiplechoice questions, thereby making success in school dependent upon memorization of facts. Even outside of school, in the workplace, companies train employees to do their jobs and then worry how to assess what the employees have learned. The need to assess has focused everyone on things that are assessable. Thus, facts have become “the currency of the educated” because they are so easy to measure.

The problem with all this is twofold. When our institutions of learning focus on test results, it follows that they need to focus on teaching what is testable. This leads to throwing out the baby with the bath water. The question of what to teach gets perverted by the measurements that are already in place, thus making curriculum change impossible. But, perhaps more importantly, there is a second problem that revolves around the issue of our understanding who we are and what makes us tick. As long as we understand ourselves to be a collection of conscious knowledge that we can recite back on demand, we lose an understanding of how we work and of what mental processing is all about. What we think we know may not be the same as what we actually know.

Most of the really important knowledge people have that enables them to do things and perform and behave in their daily lives is not consciously known to them. People perform most of their daily functions in a mindless manner, knowing how to do things without understanding why they know how to do them. Even so, the current educational system says that students must learn to talk about what they might do, rather than actually doing it. When conscious knowledge is portrayed as the sine qua non of humanity or of intelligence, non-conscious knowledge is relegated to a nether world of knowledge not worth teaching. But as long as we only teach what we explicitly know, because this is, in essence, what we think of as intelligence, our educational systems will fail.

People have some profound misconceptions about what it means to know. Those misconceptions come about because facts are what people recognize as what they “know,” but we don’t really know what we know. We don’t know how we learn, how we understand, how we come to feel what we feel or believe what we believe. In fact, all we know for sure are our own sensations, our own thoughts (to a limited extent), and facts.

For as long as there have been humans, they have had to learn where they were going, how to perform actions of various sorts, how to interact and get along with other people, how to communicate, how to reason, and so on, without “knowing” any of these things. That is, they can do many things without knowing how they do any of them. The knowledge by which they do these things is simply not conscious. Writers, teachers, preachers, counselors, and others often attempt to make such knowledge conscious. They tell people explicitly what they may or may not know implicitly. Sometimes learning something explicitly in this way, thereby making previously non-conscious knowledge conscious, is helpful, and sometimes it is not. The issue is whether one can convert conscious knowledge into non-conscious knowledge. If knowledge is only conscious, it is not of much use.

This is what we mean by consciousness: the explicit knowing of what we know or think. This assumption is the basis of our current instructional system and our whole conception of learning. And, while there is truth in the idea that we know a great deal that is quite conscious, the amount we “know” that is non-con- scious is far more extensive and more significant. We often cannot talk of actually knowing this stuff, however, as the whole idea of “knowing” in this sense is an idea rooted in our conception of consciousness and has nothing to do with non-conscious thinking.

But it is this nonconscious knowledge that we want to teach in our courses because we want our students to be able to do things—not just regurgitate facts. The art of educational design depends upon knowing the difference between what someone tells you that wants to know and what you know they need to know in order to do what the knowledge that was prescribed was to supposed to enable them to do in the first place.

What should the Virtual University teach? Typically, when people talk about building a university in a new way, they find that they cannot fight the prevailing view that the subjects that must be covered are the ones that professors have always taught. The idea is that these subjects cover the breadth of human wisdom and college is a sort of foray into what is known about the world. The idea of a VU is, today at least, an idea about new delivery methods, and is likely to evoke as much interest amongst those who are serious about education as correspondence schools have. We assume that VU means “same old subjects, new delivery method.”

VUs exist today. We know what subjects they teach. Today, VUs offer what can be offered at a distance. The courses themselves are usually watered-down versions of everyday college courses, and are, at their best, almost the same as everyday college courses. It is no wonder that the idea of the VU has not sparked all that much interest so far; it just seems like college without the good stuff. If it’s the same old stuff in a new bottle, then we can be sure that much has been lost because of the lack of personal interaction and that very little has been gained. If it is different stuff, then we need to ask what will be different, what the value of such a difference might be, and how these differences will come about.

But there is no reason that the VUs of the future should be watered-down versions of State U. We have the opportunity here to ask some hard questions about the fundamental nature of education, and to build better VUs as a result. Courses as we now know them won’t remain, nor testing as we now know it. VU will cause a lot of assumptions about school to change, and one of the biggest is the one that says students can only take courses at set times and for established periods of time, with tests at the end to measure what’s been learned. The whole question of measuring student achievement, as well as measuring the success of the courses themselves will, by necessity, be answered in new and better ways.

Certain basic assumptions about what we should teach have held sway for decades. These assumptions are based, typically, on what we have already experienced in school. We believe that there should be a lot of math, and some science, and some literature, not because some esteemed body of experts thought this out one day, but simply because it has always been like that. In short, most people believe that we, as a society, should be teaching what we have been teaching. We never hear anyone ask why we teach math, but we hear all the time about how math test scores are down and why this means that we are in trouble as a nation. Perhaps we should teach math, but the reasons offered (it helps in reasoning, it provides discipline) always seem to be kneejerk reactions rather than well thought-out responses. But in VU we get to ask the question: what would happen if math teaching were embedded in courses where it was needed, rather than taught independently? The same question should be posed for any subject being taught at VU (or any school, for that matter).

In fact, each and every subject can, and ought, to be questioned. There is no way that State U will change what it teaches in any radical way in the near future. There are too many vested interests. Teachers don’t want to have to give up their livelihoods or have to learn something new. In general, teachers think that schools should offer exactly the subjects that they know how to teach. Whether or not it was a good idea to teach math in the first place, there is no getting rid of it now, as there are all those tenured math teachers, all those book publishers who have invested in expensive textbooks, all those test providers, and an entire nation that is convinced that if they went through all that pain it must have been for a very good purpose.

A VU that starts by refusing to offer mathematics would be rejected immediately as not being serious. But VU will have to respond to the needs of its students, so perhaps we should begin to consider what those needs might be and where those needs come from.

If we want to see the VU of the future today, we need to look at the large corporations who have come to understand that they need to train their employees in a variety of skills right now. More and more, corporations are taking it upon themselves to open large campuses where they teach new employees the business skills they will need on the job. The expenditures are so great for these “corporate universities” that they are beginning to deliver courses virtually in an effort to save money. They have found that, while saving money, the virtual courses have the added bonus of allowing practice in realistic situations. So, the virtual courses, like the air flight simulators that were their progenitors, can be better than the real thing. Corporations using this model are quite willing, even anxious, to hire experts from outside to teach skills that the corporation does not have its own experts for. Corporations also worry that they will lose the experts they do have, and so they have begun to immortalize those experts in software. Further, they have come to understand that many of the skills their employees need to know are not particular to any one company. Negotiation is negotiation, for example, and it can be taught by a variety of negotiation experts.

But even as corporate universities are growing and enjoying some successes, they are under attack. The CEOs of these companies want to know if the corporate university is producing results. They are spending a lot of money in order to earn more. How can they be sure that this will happen? One way to help with this is to farm out certain courses, the ones that don’t require instruction from someone specific to the sponsoring corporation. For instance, if the corporate university has to teach accounting principles to its staff, does its form of accounting vary from those principles used in another company? They shouldn’t. So why can’t the company “outsource” the teaching of “accounting principles for non-accountants?”

One reason is that it isn’t easy to find such a course. Companies located in a big city could band together to create such a course, but that kind of cooperation in training departments of corporations is unlikely. Alternatively, assuming there is a major university nearby, they could send their employees to that course. But that course probably meets only three hours a week over many weeks and the company wants its employees to spend all day every day for just five days on that subject and then get back to work. And most importantly, major universities typically don’t teach accounting for non-accountants.

However, because VU will be constructed as a for-profit entity, it will be paying a great deal of attention to its customers (unlike its non-profit competitors). So, it will have that course in accounting. Further, that course will be taught by the best accounting teachers in the world (or at least the subset of them that think that building such a course would be a good idea). That course would be a learn-by-doing course. No listening to dry lectures about accounting. Rather, problems would be presented, solutions worked out, and commentary on those solutions with help readily available. Such a course could be built in two ways. One way would be to have live helpers available on-line to teach students as they had difficulties and another would be to have that help totally automated. Both of these solutions are possible. The experts would appear in video as needed and would tell interesting stories about their experiences as a way of helping guide the student to do tasks that might be representative of those they might have to do in their actual work. Presumably, in a course such as this, this would mean having students use their knowledge to allow them to read reports that contained accounting information rather than having them do the accounting themselves.

This last point is a rather important one and may help us begin to understand what tomorrow’s VU might look like. In today’s universities, professors of accounting teach accounting courses. This seems natural enough. However, there is a disadvantage in having professors who really know their subject matter be the teachers of such courses. Professors like to show off their expertise. Or, to put this in another way, they see, and love, the theory that underlies their work. They want others to see and love that theory as well. They cannot help but believe that if only students would be exposed to the real principles behind their work, they too might love it and some of them might change their lives and go into that line of work as well. This is a professor’s dream. It is also a student’s nightmare.

Most students are taking a course because they have to, or because there is something in that course that they think they might need to know, or maybe simply because they think they might love that subject. This latter state of affairs is usually the rarest and, when it comes to corporate courses, is more or less nonexistent. So, the professor who loves his subject is often better replaced by a practitioner who understands the use of that subject. (This happens from time to time in universities, but mostly in foreign language courses where it is possible to hire native speakers of the language in question who can help students learn to speak and who are not, themselves, usually scholars of the language.)

How will VU change and improve on this? VU could afford to create the special courses needed by corporations or desired by business students. They will be taught by many experts who are funny and interesting, to boot. These experts will teach through example, examples taken from real problems in the life of an accountant or an economist or a sales manager. Courses will be structured around the needs of the students. Thus, the age of the math course that is taught by the math professor will give way to the era in which courses are designed to teach math skills people might need in real life and are taught by one or more people who know what real life looks like.

So, accounting for non-accountants isn’t likely to exist at State U, but is likely to exist at VU. Moreover, to the extent that this course does exist at State U, it will have the wrong teacher, while the VU course will have exactly the right teacher (or teachers). Third, the VU course is likely to be more interactive, more fun, more problem-oriented, and more practical than its State U equivalent, if there is a State U equivalent. VU is going to win every time.

Will Corporate U decide not to teach accounting for non-accountants if the VU version is really first-rate? You bet they will. Every course they don’t teach themselves saves a great deal of money. Not only are we talking about the cost of instructors and the cost of classrooms, we are also talking about the cost of airfares to fly students to corporate training centers and the cost of hotel rooms and the cost of down time for employees who are in class and therefore not working but still getting paid. We know this is true. In the first virtual business course built at the Institute for the Learning Sciences at Northwestern University and delivered to Andersen Consulting (AC), designed to teach basic business practices to new hires at AC, the company estimated that they saved ten million dollars a year by using it!

So, it is a safe assumption that VU will begin by offering stuff that is not offered by others or stuff that is best offered virtually. VUs will also provide specialized courses beyond those needed by businesses. One can expect that courses that the general public wants to take, typically offered in schools of continuing education, will become available virtually; courses in foreign languages are a good example. But the real question is not which courses come first, but with the question of courses themselves. We need to ask what courses are intended to provide and what they actually do provide. VU may be able to make significant impact by providing what people need rather than what people have been taught to want.

We tend to imagine, when we get to school, that we are there to get specific explicit knowledge. All that we have been told about schooling and education has been that the guy who knows the most stuff wins. So, schooling of most any sort tends to make knowledge explicit. Much of what science is about, for example, is the attempt to make explicit what we know so that we can make better assumptions about the world. We may all know that objects fall when released from our hand, but science means finding out the general principles behind that fall and making them as explicit as possible. The process of making knowledge explicit is an important basis of intellectual life.

The problem is that there is a difference between discovery and memorization. Making concepts like gravity explicit is what science is all about. Memorizing the rules of gravity does not make one a science student, however. Training to be a science student would presumably entail learning about gravity, but the intent of that instruction should be to teach one how to make generalizations based on data gleaned from evidence. Real science training is like this. In graduate programs in science, students are trained to behave like scientists by allowing them to do science, both in the laboratory and by way of theoretical conversations and written papers about ideas.

But what takes place in science courses for undergraduates is not about training scientists at all. Courses like “rocks for jocks” and other science courses intended to cover a certain amount of material are basically “feel good” courses. Someone has decided that every student should know some science and a requirement has been established. The intent here is to make everyone feel as if they are learning science. But professors usually hate teaching these courses and students usually hate taking them. Nevertheless, take them they do, and so, there must be a test. There must be a test to see if anyone was paying any attention at all. Otherwise, students could fake it by simply showing up every day and paying no attention whatsoever.

Of course, in fact what is going on is that students are faking it. They are faking it by figuring ways to pass the test without paying too much attention. Professors need to teach in order to get paid and students need to get grades in order to get onto the next professional plateau, so everyone conspires to make believe this is actually education. Unfortunately it is not.

Here again, VU can come to the rescue. The first step in this regard will be to ask what a good science course might entail. The second step will be to ask who would take such a course and why. Certainly any good course in science should entail doing science. This might mean a range of things from working in a laboratory, to performing experiments, to writing theoretical treatises, to trying to apply scientific reasoning to everyday problems. Whatever it means, students would be doing science, not memorizing facts about science.

This is a fundamental issue for the VU. Can it provide courses that offer what students need rather than what the educational system has made us all believe we need? VU should offer the chance to do science, not to hear about it. If it can do that, it will succeed, otherwise it will be just another delivery method.

We tend to romanticize college. We form attachments to our school and we are sure it is better than all the others. Yale students chant “Harvard sucks” at football games and they mean it. Apart from the loss of these pleasantries, we can ask if one VU will be better than another. The answer should be that this will become a meaningless question over time.

Students would be free to take courses from any source. Why would any one university be the sole dispenser of courses or of degrees? In a physical world, unless you are willing to “go away to college” you are bound to attend the school that is nearby. Shopping around for the best courses is not really feasible. In a virtual world, the only thing that would prevent students from looking around for the best courses would be the inability or unwillingness of the institutions themselves to cooperate with each other. If there are many VUs and one has a really good science course and another has a really good business course, shouldn’t a student be allowed to take both? After all, issues of location are over. In principle, one could always take a great business course at Stanford and a great physics course at Harvard and each would give credit for the other’s course. Given that the VUs will be for-profit enterprises, one can expect that this cooperation will take a while to happen, but happen it will. Free market forces will prevail, and no quality institution will legitimately be able to deny a student access to a quality course.

This is true today, of course. Any Yale student could go hang out at Harvard, take some courses and expect Yale to give him or her credit for those courses. For the most part, this would not be a problem at Yale, except they would begin to worry about whether they were getting enough of the tuition money to justify giving out a Yale degree. Make no mistake: that, not the quality of education, would be the issue.

At VU, one doesn’t have to relocate to Boston to take a course. So, we would expect each student to take the best courses available. A VU that refused to grant credit for good courses taken elsewhere would have to be the very best or it wouldn’t survive. It is my guess that VU will offer courses from everywhere provided by many different providers. The job of the administrators of VU will be determining which courses are worth giving credit for. So, evaluating the quality of courses will become a very important issue, much more important than is now the case in transferring credit between schools.

The courses in the real world of school and of business training will be good, high quality, fun, and goal-directed. Students will learn to do things in these courses that will help them with their own goals, goals for doing their jobs better or understanding history better or comprehending complex scientific theories. Students will remember what they’ve learned and will be able to apply their newfound knowledge in a variety of ways.

We remember in only two ways. First we remember something we have ourselves experienced. In order to learn from the experience of others we need to be having a problem in doing something for which the experience of another will be helpful immediately. Doing something right once doesn’t fix it in our memories. We must do it over and over again. The key is doing. Courses can be redesigned to allow exactly this to happen. The way to make it happen is by using goal-based scenarios (GBS).

Goal-based scenarios are not defined by length of time, number of lectures, or any other passive measure. They are defined by tasks accomplished, no matter how long they may take a particular student to do. A GBS contains a clear goal to be accomplished, helps the student play a role in a realistic situation where he may accomplish that goal, provides material to be learned to help in accomplishing that goal, and provides instruction in the form of well told stories by experts at the point when they are needed by the student to help him accomplish his goal or at the point where the student becomes curious about something.

GBSs have no time length. Some people take a long time to learn to drive well and others are good immediately. GBSs are subjectively measured in the case where the skill is one like driving and are objectively measured when the skill that can be precisely defined like hitting the target consistently with a bow and arrow. Measurement does not drive instruction in a GBS. There is no arguing about grades because there are no grades. There are simply merit badges like the Boy Scouts. You demonstrate that you can do something and you are certified as being able to do it.

Good software can be created that allows students to augment their memories through experience that relates to prior experience.

Good multimedia education only works if it mimics the human learning process. Thus, it must induce failure, allow explanation, and interest us in the stories of experts that have been made available in the GBS. This requires good simulations of real world situations. A good simulation makes the user feel that what he is experiencing is life-like enough to make the process of using the simulation educationally valid. A simulation must look and feel like some real life situation.

An effective simulation must allow for every possible path that a user might consider taking; if the simulation only allows one or two answers it in effect tells the user what to select and thus does not cause the user to believe that he is in charge of the situation and is really making his own choices. Allowing every possible choice and including real responses to each of these choices is the hallmark of a good simulation. A good simulation motivates the user to do more learning by sweeping him into the metaphor so that he believes himself to be doing something real. To construct such a program requires all possibilities in a given situations to be programmed into the simulation.

Building GBSs is not easy. Finding out what we know about a given situation is harder than it looks. We don’t only need to know the right thing to do, we must also know all the wrong things that have ever been done and what the consequences of doing them are. Wisdom is understanding what works and what doesn’t and why. We can get content by interviewing all the people we can find who have ever had experiences that relate to what the student is trying to do in the GBS. We then must link each of the stories we get from these experts into the various choice points in the simulation. A good simulation has hundreds more stories than a student will ever hear because no two students make the same mistakes. Each mistake should be connected to a story from others who have made similar mistakes. This enables the student to reflect on what he has done and to create an explanation for himself to use as a guide in the future.

The most difficult issue to tackle in changing the nature of the education system is enabling people to see the doing in what they’re learning. We are so bought into the idea that a course in a university is about the exchange of ideas or listening to and understanding the works of great minds, or getting prepared in the a basic grounding in a field, that we fight attempts to see doing where there now is listening.

Objections come from every corner: from those who have attended the lectures of great scientists and found them thrilling to those who sat in a seminar discussing literature with other students and came away with a new appreciation for argumentation and writing. And, surely, these can be great experiences and one can learn from them. Nevertheless, it is also the case that these great experiences can be simulated on line. Further, better experiences can be created. And, perhaps most importantly, we need to remember that replicating great experiences in software makes them available to a much wider range of people. Not everyone got to attend that great Harvard seminar in 1978.

My point is simply that we need to rethink the concept of a course. We need to rethink it because we can do better but we also need to rethink it because the marketplace demands it. The change in how education is delivered will happen whether we hold on to our cherished memories of a great school experience or if we vilify every teacher we ever had. It really doesn’t matter, because the students will vote with their feet (or their PCs, in this case.)

Further, we need to rethink the idea of a required course. Today universities and adult education schools are horses of a very different color. Many universities run continuing education schools that specialize in subjects that adults feel they need to know, such as computer programming or foreign languages. Adult education courses exist in a variety of venues that teach photography, ballroom dancing, investing, and so on. What can we learn from these schools? These schools teach subjects that students want to learn. There are no required courses because there are no degrees and there is a no board or faculty determining what an education in these subjects might mean. Students are in charge of their own education.

Now here comes VU. What does VU do? It really can’t set up required courses. The reason is simple enough. Even if one wanted to set up requirements, unless the VU that did that was simply the most prestigious VU, there would be others whose requirements would be simpler or non-existent. Since transferability of credit is a sine qua non of the whole concept of VU (there is no reason why all of one’s courses need be taken from the same place) no one VU could require things that others did not.

At State U, committees of professors protecting various vested interests and using some conception of what the educated person should know make arbitrary decisions about what should be taken for a student to be certified. At VU, which presumably would not have an actual faculty, this would be handled differently. It should be a consumer choice, after all. We would expect students to want to learn certain things and we would expect VU to provide a variety of offerings and to make suggestions. The best science courses would be determined, presumably, by the students who would, one would assume, vote with their feet. And, we can assume, they would vote for courses that actually taught them to do things they wanted to do.

All this change at VU will require some conceptual change in the student population. We are so used to studying, memorizing, passing tests, and then quickly forgetting what we memorized, that we have no other model of education to fall back on. We use learning-by-doing in our daily lives, but we are not used to going to school to do so. So, we may feel shortchanged at VU, wondering what we learned. We may not be able to say what we learned after all.

If one believes in a learn-by-doing model, then one has to accept, at least implicitly, the primacy of non-conscious knowledge. We really don’t know what we know. We can conduct an interview, manage employees, deal with personal problems, drive a car, hit a golf ball, and so on without explicitly knowing how much pressure is exerted when one swings or how one decides to challenge an employee or support him when he is screwing up. We learn these things by doing them and we cannot necessarily articulate them.

We will learn how to do these things and other doing kinds of things at VU. We will be able to show off what we learned when the time to do so arises; not in tests, but in life.

This brings up the issue of how success in a course should be determined at VU. It is clear that in a learning-by-doing course, the ability to do should be the ultimate determinant of success. My favorite analogy here is always the driver’s license. If you can drive, you get one. The silly multiple-choice test associated with the driver’s exam is just that—silly. The real question is can you drive? And, once it has been determined that you can, you get a license. Not a license that says you got 92%. No grade is attached. It is just a yes-or-no affair. And so it would be in any doing course. The question, is can the student do what we asked him to do? If he can’t, he can keep trying until he can.

We can certify those who have performed well in the simulation. Our job is to build simulations well enough so that they include a variety of situations that we feel are important enough to have been the kinds of things we would have tested for in the multiple choice test. If we want to know if a student understands the meaning of a certain sign, for example, we need only have that sign appear on the simulated roadway.

Despite talk about education reform, it is not at all clear that anyone wants reform if it chips away at any sacred beliefs. One of the most important of these sacred beliefs comes from scientists and scholars, and it must be carefully examined, as the strength of this belief is so pervasive. It is this belief that permeates our thinking about education, and causes even the most enlightened advocates of change to balk at some of the proposals made here. So we need to address it before moving on.

Can we redesign courses in principle from what they have been (entailing very little actual doing on the part of the student) to what they should be? Can we prescribe rules that dictate what a good course should entail? The very idea of a course may well be wrongheaded. Change will come because students demand it. VU will be the instrument of that change.