and there must have been a very elaborate safeguard put on that thing, because it was in the intelligence area-it indicates to me that this can be done fairly easily by a determined person who has the proper knowledge. And I wonder if you have any thoughts on whether it is going to be a very difficult task to guard the privacy of the information which is transmitted on these so-called dedicated lines or whatever manner you referred to them.

Mr. DIAL. I think it is going to vary considerably according to what is being carried on the line. I think the privacy has its value, which is often, at least in a municipal or governmental level, translatable into costs or overhead of the system. There can come a point where you become so involved in protecting the system that the system is no longer economically feasible, or viable.

Mr. CORNISH. That is, I might say, one of the things that bothers me because I think one of the crucial elements involved here is that if there is any surrendering of privacy that it be done on an informed consent basis. I might have no objection at all to having certain aspects of my privacy invaded if the purpose is a good one and is going to help society, but I want to give permission for my particular information to be utilized for that purpose.

Mr. DIAL. Let's examine that for a moment. You are concerned about your security at home and you might well consider how much money you invest in locking the doors, the quality of the security you are actually given. In other words, you are accepting a risk to your privacy even at home, and you are translating it into the dollars you are willing to spend on it. After your home is penetrated, you may be willing to spend additional dollars to get more protection.

There is a second consideration here. In moving into a computerized system, you need to compare the privacy that that system gives as compared to the privacy of manual files, which is the system that predated it. How much concern did you have for privacy then? How much money did you spend on the protection of those files? True, they were immediately readable-you didn't need a machine to read them."

Mr. CORNISH. Well, I think this naturally brings out a very wide, general thought to mind. I am sure in all of the development of this technology that a certain amount of cultural lag is involved here, and one of my concerns is that the social scientists should be working hand in hand with the technologists to deal with some of these other questions like privacy, which is a good example, and there may be many others, because it can change the whole character of American life styles. Do you personally feel that there is enough exchange of information and cooperation between the social scientists and the technologists?

Mr. DIAL. NO.

Mr. CORNISH. In this area?

Mr. DIAL. No; there has not been. The best evidence of this is that technologists are not very sensitive to the question of privacy unless they happen to be technologists of privacy. We had a very difficult time early in the USAC program in persuading not only the technologists, the systems people, but in persuading the city administrators, the city councilmen as to the sensitivity of this question of privacy, and how, if machinery weren't set in motion to deal with it in

a way that would gain the approval of the community as a whole, the entire program could be defeated. I would suggest to you that it was somewhat difficult to communicate our concerns. So difficult in fact that some coercive measures were applied in order to see that there was the requisite level of attention.

Subsequently to that, we had a great deal of written communication as to the problem of privacy and how best to honor it and the problem of security and how best to provide it. I think at this point we are really achieving what we set out to achieve, but it was a hard thing to turn around the initial reactions. So I think you are probably right, Mr. Cornish, that there needs to be a lot more communication between social scientists and technologists.

Mr. CORNISH. Let me ask one final question and then I will step -aside, and that is throughout these hearings, I have heard about the importance of innerconnection of systems and so forth. This appears for many reasons to be very desirable, a very desirable thing so that you don't have too many different systems operating and they can then tap into each other. Also there would be a great deal of economy and efficiency.

Mr. DIAL. System compatability.

Mr. CORNISH. Now, looking down the road-and I am raising this specter of privacy again, or what have you-but where do we end up? Do we end up with a national data bank? Is that where we are going to end up?

Mr. DIAL. No; I don't think so. Very often the people at the top of a system can get the best information out of that system if they also have information from another system with which to correlate and a need to go back to some earlier unit of aggregation. It may be at the individual level, it can be at the block level, the community level, State or what have you, but this will be necessary for them to make informed judgments about the effectiveness of one program or another. This can be done and the identifiers stripped away at the source or the information aggregated and hence reported in an aggregated form for the block level, the city level, what have you, so as to remove any question of the problem with privacy with respect to the individual at least.

We find that individual privacy is not always the greatest concern, however, even where it is asserted to be the greatest concern. For example, take a hospital information system. There is a great reluctance to participate in a chain of hospital information systems in a health information system, because even if you strip away personal identifiers, it still becomes possible to come back and identify the doctor, perhaps the mortality rate of his patients in surgery, or the economy of operation of the hospital, and there are a great many things that you could learn about that institution, so we are concerned about, not only about the privacy of the individual, but the privacy of the institution too. This may be a legitimate concern on their part. I don't know. But it is something that a community can decide, and I think that community, the local community, is probably in the best position to say what is an acceptable level of disclosure, whether with respect to a person or an institution.

Mr. CORNISH. I find almost every piece of paper I fill out today asks me one question which is a thread that runs through on every form, and that is they want to know my social security number. I have a feeling that there is a reason for that, and it is an uneasy feeling because is that going to be, you know, the dossier of the future by social security number?

Mr. DIAL. I don't think so. It is a personal identifier and a data access control board would have to think very carefully whether or not that serial number was to be made available with the information.

Mr. CORNISH. I might say a few years back, you may have heard about this case, but some fellow out in California was asked this question so many times that he decided to change his name legally to his social security number.

Mr. DORDICK. One brief response to your first question concerning the interaction. Last year there were attempts made by the Office of Telecommunications Policy to embark on a set of pilot projects in various cities to test the capability of broadband communications. If I remember correctly, many of the proposals were perhaps as they should be in the main, very technological. They were testing hardware. They were testing the ability of the systems to work. I would like to recommend very, very strongly that if ever those pilot projects are done, there be a very strong social science input because not only do we test how the hardware works, we have to test whether the people work with the hardware and the hardware with the people. I would strongly urge whenever these pilot projects are done, that the social scientist be a major contributor.

Mr. SMITH. One more dimension that might be added to this whole question of privacy is that of interest. The clientele for the privacy issue is shifting and deepening. Long ago, privacy was sort of an elitist concern of the American Civil Liberties Union and others. It was, I think, to some extent scorned by many people who might be called the average American voter. The idea of privacy during, let's say, the McCarthy era, in regard to his investigations, would be simply incomprehensible. I don't know if you could get one person in 10 on the privacy side then.

Now, however, interesting things are occurring. For example, I have had a number of conversations with community groups in large cities, black community organizations, and so on. They tell me when they go into black areas of the cities to extoll the wonders, so to speak, of cable television, they come up with a solid wall of opposition. The blacks and the poor are sure that this is another device that "the Man" intends to use to discover what they are up to. They think this is a plot of the higher powers that be to encroach upon their privacy and to the extent it is based on myth, and to the extent this feeling is based on real problems we have mentioned today, is not very clear to these people. What is clear is that privacy is becoming more important and the nature of privacy as an issue, if it is an issue, is reaching down into our society right now.

Mr. CORNISH. I might say the framers of the Constitution had a very clear understanding of what privacy meant and I am glad they dealt with it.

Mr. SMITH. One has to be grateful for many things that they did. Mr. DIAL. Well stated.

Mr. MOORHEAD. At this point, I would like to ask unanimous consent to include various studies, articles, and papers and miscellaneously related data in the hearing record. Without objection, so ordered.

[See appendix for material referred to above.]

Mr. MOORHEAD. I think our time is up. The bells have rung and I must go over to the House floor and vote.

I again want to thank you all for giving us the benefit of your expertise and your candid and helpful thoughts on this subject. I don't think the broad implications of electronic information technology has been adequately covered or considered up here on Capitol Hill. We are making a stab at it. Thanks to you and other witnesses, we hope we can shed some valuable light on the subject.

The subcommittee is now adjourned.

[Whereupon, at 4:30 p.m., the subcommittee adjourned, to reconvene subject to the call of the Chair.]

APPENDIX

ADDITIONAL MATERIAL AND CORRESPONDENCE RELATIVE
TO THE HEARINGS

THE LIBRARY OF CONGRESS

CONGRESSIONAL RESEARCH SERVICE

BRIEF SUMMARIES OF MAJOR TELECOMMUNICATIONS SYSTEMS

(By Charles E. Bosley, consultant, Government Information Systems)

This paper briefly describes characteristics of major telecommunications systems, with particular reference to the way in which different systems employ new technology and new devices.

Information has been compiled from a number of sources which are listed in an appendix.

COMPARISON OF SYSTEMS

There are four elements to be considered in comparing telecommunications systems: terminal equipment, local distribution facilities, switching centers, and long-haul trunks. All systems contain the first two, although they do not necessarily have all four.

In another aspect, all systems utilize either wires or wireless broadcasting, or some combination of both. Communications signals must be transmitted over wires or through the air.

One convenient way to generally categorize major telecommunications systems is by distinguishing between those that utilize wires or cables for local distribution and others that utilize wireless broadcasting for local distribution. This distinction is recognized in the listing that follows. Systems with wired distribution facilities—telephone, community antenna TV, and broadband cable— are described first. Following are those with wireless distribution facilities— radio and TV, mobile radio and microwave.

TELEPHONE SYSTEMS

Telephone systems provide the major vehicle for personal communications. There were 125 million instruments in 93 percent of all homes in the country by 1971.

Terminals are telephones, of course, but may also include teletype machines and a variety of other equipment.

Local distribution is by pairs of copper wires called loops which connect subscribers to local switching offices. Each subscriber has an individual channel into his home from the switching office. The cost of the system is high because a large number of separate channels are required.

Local switchboards are connected to toll offices or exchanges via multipaired cables-bundles of wires. Toll offices are interconnected by trunks which are either coaxial cables-clusters of copper tubes containing insulated center conductors or wireless microwave relays.

Telephone systems were originally intended to provide only voice communication. Copper wire loops could only carry low frequency transmissions, but that was all that was needed to reproduce the sound of the human voice.

However, as advancing technology produced new communications devices, telephone systems were asked to accommodate more and more kinds of terminals and higher and higher frequency transmissions. Extra telephone lines-extra wire pairs are often required to serve new terminals. Modifications in both switching equipment and long-haul trunks are often necessary. Costs increase.