with the National Science Foundation grants where mathematics and reading from kindergarten up through the sixth grade level will be tested and evaluated on a very formal basis by a nonparticipating or nonbiased party, so to speak.

We have also been testing in the colleges, and we continue to test every day to see how well we are doing. I don't think we have ever found a failure; even the worst written material seems to do as well as the classroom. And it gets much better usually in saving students time as they proceed.

You understand it's very difficult to compare it to what one would call standard educational measurements, because most people don't like to measure, and they haven't been measuring in our standard educational system.

The only measure that the State imposes upon a school district as to whether they will get their funds and how much in any given year is the average daily attendance. And so that is what they measure.

There are other tests that are given by the better schools. I recently noticed in one major metropolitan school system that for a while the school board was forbidding any standard examination; that is one way of making sure there's no comparison.

With a Plato-like system it is clear for the first time whether we are teaching successfully or not. You cannot sweep the results under the carpet and send the child home. The results are recorded in the system and we are forced to make the measurement whether we want to or not. And we will make that measurement, and in the next few years, we're going to be able to answer how well we are doing in much more firm terms than I did just now.

Mr. ERLENBORN. Might I ask a question? Is there any component of your measurement to measure the emotional development of the child? I have in mind really that a primary source of education is in the home. I think the formal education is only a supplement to a great extent.

The child learns an awful lot more than just information, the mores and-well, the value of an example is so important. The child tries to emulate his father or mother or the teacher or one of his contemporaries in the classroom.

Are we in the danger of this being the tendency of human nature, of the child to start emulating the computer that he's dealing with? Mr. BITZER. I haven't seen that. Although more specifically.

Mr. ERLENBORN. Are there any measurements or studies being done on it?

Mr. BITZER. We've done attitudinal studies related to the type of teaching, because we were interested in knowing not only whether we taught successfully, but whether students felt it was a good mode of learning, whether they learned fast, slow, or so forth.

The attitudinal measurements overwhelmingly supported Plato as being the best teacher in the university, second only to one human teacher who happened to be one of the teachers we had here this morning, a classics teacher. So attitudinal exams indicated as a teaching media, they were satisfied, but that doesn't answer your question. specifically.

One of the benefits that you get from this type of teaching is the speedup. It takes about one-third to one-half the time to cover the same material.

Now, contrary to what you might think with machines being very depersonalizing and cutting down on the human contact examples and discussions and so forth, it is just the opposite. Now, the teacher really has time, twice as much time, as before to pay attention to the students. It is not an all-day hand-holding task in the classroom, because the student is being taken care of until he needs help. Now, the teacher can spend more time with the students who need help, and there is more time, on the average, left over for all students to interact together.

So the human contact really goes up. If you ever visit one of these classrooms where Plato teaching is going on, it is a beehive of activity; and those who are concentrating on a lesson don't get distracted; the other students are talking to each other and helping each other. The social function of the school goes on at even a higher level, I would say. I hope that at a later date we'll be able to measure some of these items that you have mentioned.

In educational measurement, there's what I call the "uncertainty principle," similar to the one in physics where the product of two variables remains constant. The harder something is to measure, the more likely it is that it will be valuable. So if you can measure it very easily, the chances are it doesn't really measure the kinds of things that will determine quality of citizenship or how good a student will be in a particular field.

Useful measures of these qualities are difficult to make, which corresponds to the fact that they are probably very valuable. And so, the things you've asked about are valuable measurements. It's going to take a long time; it's going to be very difficult. But we are building a system to help us carry out some of those measurements.

Mr. ERLENBORN. Mr. Copenhaver, have any questions? Other members of the staff?

Mr. Cornish?

Mr. CORNISH. Yes. Thank you, Mr. Chairman.

Dr. Fano, I was thinking of one of the statements you made in your presentation that a computer communication network must ultimately appear to its users as a single system. Then you went on a little later to say this does not imply that all information must be owned or operated by a single organization, but it does imply that all such networks must be interconnected, and their users must be largely unaffected by the internal structure of the overall system.

Now, in effect, are you suggesting that we establish what might be characterized as a national data bank?

Dr. FANO. No, sir. One can talk about this at different levels. I am talking about an integrated information network. Now, of course, data must be there, but a lot more than data must be there to make it valuable.

Second, while there may be in that system information about people, most of the information will not be about people. So we are talking about a different sort of thing.

Furthermore, as I mentioned before, I am highly concerned about the matter of individual privacy, and while I am talking about this as a need for the future, I would not want to see it happen until we have developed adequate means for protecting individual privacy;

and furthermore, legislation to that effect. I don't see how we can meaningfully develop the technology to protect individual privacy out of the context of, let me call it, some sort of social view of how to handle the problem of individual privacy, because it is an extremely complex problem.

As of now, I see almost no progress being made in understanding the nature of the problem. Some legal inventions are badly needed to make sense out of the situation.

For instance, I was chairman of a committee at MIT to study the problem of protecting individual privacy, and we ran into difficulties for which we couldn't find any good solutions. One of them is the fact that there is very little information that is private to only one individual. In most situations, there are two or more people involved.

That is, almost always you have conflicts of privacy; that is, in order to protect one person's privacy, you may have to violate another person's privacy. And how to resolve this question is very complex.

I feel that many of those conflicts could be eliminated by wiser handling of information. But some of them, I just don't know how they could be resolved.

As you know at present, even the concept of privacy is not weli understood. I might be wrong in this, but I know of only one man in the legal profession who is trying to really study this concept, and to invent ways of dealing appropriately with privacy problems. Mr. CORNISH. Are you referring to Dr. Westin?

Dr. FANO. No; I'm referring to Prof. Arthur Miller. I know both of them. Professor Westin is largely concerned with the sociological aspects. Professor Miller is more interested in the legal structure, and the regulatory aspects.

Professor Miller used to be at the University of Michigan. He is now at Harvard. I got to know him better since he moved closer.

But these are big problems that I believe are going to become roadblocks in the movement that we have been talking about unless they are properly handled in time. The whole question of management of information is just not understood.

Mr. CORNISH. Do you have any suggestions on the way that social scientists somehow can be brought into this picture with the computer specialists-the technologists-to work hand in hand on some of these problems? I noticed in response to some of the questions, some of the panelists-and I don't mean this to be critical-you said well, this isn't my field, you know, in dealing in questions of housing or in dealing with other questions.

That might have very sensitive spinoffs, sociologically and politically in this country in the use of such technological systems.

Dr. FANO. I wish I had an answer. At the moment, I can only speak about my personal frustration. For a number of years, I tried to get people to interact in many meetings. But there are various problems. One is what you may call the cultural gap. That's a very real problem. The second is that we are dealing here with interdisciplinary questions, and because of the structure of universities and of other instituitions, involvement in interdisciplinary questions is, by and large, discouraged by various facts of life, let's call them.

And yet, these questions are important. Perhaps some explicit interest and push from the Federal Government may wake up more interest. Perhaps some people feel that really these are not important questions. That is why they don't pay attention to them.

Of course, one of the difficulties is that it has to be an interdisciplinary effort. I don't think the social sciences can go on their own, and I am aware technical people can't go on their own either. They have to work together.

Mr. CORNISII. Well, I think you would agree with me that although many of these systems have a tremendous potential for good, certainly the demonstration I have seen here today indicates there is a possibility that they can be-if you'll excuse the use of the wordperverted for evil purposes, if that opportunity is taken advantage of. Dr. FANO. Well, let me speak more specifically about that. Now, let me leave for the moment, education aside, and speak of the use of computer communication in the operation of society.

I see that depending on the technology we choose to develop, there are basically two directions in which we can move. One direction, which of course I would prefer, is the direction of developing the technology to help people, as individuals, to cope more effectively with the problems that face them; thereby increasing the capabilities of the whole society. That's one direction.

The other direction, which has been the trend up to now, is to automate more and more functions in society. My feeling is that if we continue in this direction, we are bound to end up with a society of the 1984-type; that is, a society operated by a bureaucracy which maintains its control by total control of information.

And I want to stress that that can happen. And if it does happen, it is likely to happen not because of any ill intent on the part of individuals. If it does happen, it is likely to happen as the result of thousands and thousands of decisions made by well-meaning individuals that are trying to do the best they can under the circumstances. And if the technology is not right for what ought to be done there is nothing else that they can do.

So I think we have to really worry whether we move the technology in the direction of helping the individual or in that of solving the bread and butter problem now in the simplest and most economical way. That is a very serious question.

Mr. CORNISH. I must say that I share your concern.

Mr. Chairman, do I have time for just two brief questions?

Mr. ERLENBORN. If they are short, because time is running short. I'm due at my office in a couple of minutes, and we're going to have to let the witnesses go.

Mr. CORNISH. Well, we saw an interesting demonstration here this morning, and I wanted to get this on the record. Mr. Bitzer, I understand you have 250 stations on Plato.

Now, is it possible by typing the proper code on your keyboard to determine what any of those 250 stations are doing at the precise moment that you type in the code letters or code numbers?

Mr. BITZER. The answers to those questions are yes, and there is a reason for that.

Mr. CORNISH. Well, I understand there may be a reason, but I just wanted the technical capability.

Mr. BITZER. But under those conditions, the terminal which you are watching, is told so.

Mr. CORNISH. Is what?

Mr. BITZER. Is told that it's being watched and by whom.

Mr. CORNISH. Is it technically possible to tap into these telephone lines from another point, and to pick up what a station is doing?

Mr. BITZER. With the terminals the way they are now, if we were to hook two terminals on the same telephone line, they would give the same display. That can be altered if it becomes necessary.

Mr. CORNISH. So the answer is in the affirmative?

Mr. BITZER. Yes, sir.

Mr. CORNISH. Thank you.

Mr. ERLENBORN. I want to thank all of the witnesses who appeared here today. Your testimony has been most helpful. I am sorry that we have gone on so late.

I know that there are still questions that other members of the staff and I would like to ask, but I think the time has come for the subcommittee to stand adjourned. The subcommittee will stand in recess until 10 a.m., Tuesday, April 17.

[Whereupon, at 4:03 p.m., the subcommittee adjourned, to reconvene at 10 a.m., Tuesday, April 17, 1973.]

95-284-73