great that we should go ahead and do it. I am inclined to agree with this, and I am inclined to favor this type of legislation.

On the other hand, as a lay person, I feel vaguely uncomfortable. Perhaps we are creating a potential monster of some sort here. Is there any danger of getting into an authoritarian situation where a specified agency has a vested responsibility to say what is true and what is not true?

Dr. HORNIG. I understand the problem, sir. I think that the best answer to it is that there are many checks and balances built into the system. A number just doesn't appear in one of their publications as the established truth. As I said, it also appears in the original literature. The people who are using it have and can disseminate through many channels their own views if the Bureau in its publications were to be erroneous in its judgments. There are so many cross-checks or different ways in which information is both used and handled, this technical information, that I think the odds on anyone being able to make stick with the Government or anyone else an erroneous value for very long or to play good with respect to the correctness of data wouldn't stand up very long in the scientific and technical circles.

Mr. MOSHER. You don't see any danger that regardless of evidence that other people can produce, this agency will say: "Well, this is it"? Dr. HORNIG. Well, on a smaller scale this has been in operation in the past and I have seen no evidence of such a complaint in actual operation in the past.

Mr. MOSHER. What are the implications when you use the word "standard" in this bill? Is standard reference data the same as critically evaluated data?

Dr. HORNIG. Essentially. It is data which meets certain standards of quality and reliability.

Mr. DADDARIO. Will the gentleman yield?

Dr. HORNIG. It has to be critically evaluated to determine whether it meets the standards of quality.

Mr. MOSHER. But the very fact that you use the word "critically evaluated" means that your standard is at best still a matter of opinion? It may be the very best available opinion, but it is still an opinion, is that right?

Dr. HORNIG. Well, yes; it is an opinion but it may have many foundations. For example, one of the most important single constants in all of physics is the velocity of light. Now, this has been worked on for two centuries. And quite aside from books, about once every 5 years a paper appears which is the critical attempt to evaluate all the different determinations. They never quite agree. But then it isn't just a guess at the best. It is then a discussion of the probable sources of error in each of the different approaches that have been taken and an attempt to see where there might be systematic errors or where it is simply random errors. And on the basis of such a discussion, they arrive at any given time, at what is considered the most probable value in that case the velocity of light.

Now, the history of progress has been that time after time, as better measurements were later made, actually new measurements, it has turned out that the old best evaluated number was wrong to more than the estimated limits of error. So that you can never be "right,” but

you can simply get "better" in time.

Mr. MOSHER. But there will be areas, though, where it can't be determined quite so precisely as the speed of light. Won't you be publishing standards

Dr. HORNIG. Yes. That particular number is known to about eight significant figures now. But, nevertheless, I think most of the data that is compiled is data to quite a high precision, not casual determination, but not nearly as accurate as that. I think most of these data are high precision, for example the thermodynamic tables, it is mainly published to about five significant figures.

Mr. MOSHER. Will there be a problem as to how far you can go, what areas can be dealt with in this fashion and what can't be dealt with, or is that all pretty much understood and acknowledged in the scientific community?

Dr. HORNIG. I think this is part of the practical operation of the system, that is, determining the things that you compile. There are two conditions: one is that it be needed by someone, and to be useful it has to have some reasonable level of precision. If I'm going to design industrial processes or space systems or rocket motors, that determines it on the one hand. The second condition is, there is no point of setting standards beyond the best available level of precision at any point in history. So, if what is needed isn't much better than what has been already attained you don't do it at all. But if the two conditions mesh together, that is, you can arrive at something more useful than what you already have, you do it. It is a combination of the two conditions that will have to be faced.

Mr. MOSHER. They will have to face some difficult decisions, won't they?

Dr. HORNIG. Oh, yes, they have to face the decision. I mean at any given level they have to decide what is the most fruitful areas to put their effort into.

Mr. MOSHER. Thank you, Mr. Chairman.

Mr. DADDARIO. Mr. Vivian.

Mr. VIVIAN. I would like to ask a few questions. Some of these may be better answered by Dr. Hollomon at a later time. If so, please

say so.

This question has been asked before, but I want to repeat it. Why is new legislative authority required?

Dr. HORNIG. I think the main answer I would give-there are some details is that when we talk about the provision of a general service for many agencies of the Government, it becomes very hard for the agency providing it to justify it in terms of its own particular missions. And what we are talking about here is the provision of a general service for the many agencies of the Government and for the private sector. So I think that in order to make this viable in terms of future funding, that the most important thing that is involved here is the general expression of intent by Congress that this general service should be performed and not justified strictly in terms of the Department of Commerce's own needs in this case.

Mr. VIVIAN. I can understand very well why one may desire authorization authority with which to talk to the Appropriations Committee. But it seems to me that this bill does contain some sections not presently authorized.

These relate principally to the ability of commerce to restrict the use of published data and/or its ability to publish the data outside of the Government Printing Office. It seems to me that the real legislative authority granted in this bill lies in these back sections. The authority contained in the earlier sections already prevails.

I also have a question on section 7 (b) which states:

Copy any data compilation hearing the Standard Reference Data Symbol or mark adopted pursuant to section 6 of this Act.

Does that mean to copy the data compilation or to copy the mark? I can't imagine that people won't be copying the data compilation time after time. If you tried to stop them, it would be nonsensical. So it must only mean that they can't copy the mark.

Dr. HORNIG. I feel on shaky ground trying to interpret all of the meanings that might go into those words. I would personally regard it as meaningless to prohibit individuals from copying the marked data. I would assume this applies to the republication of data. If it doesn't it ought to be clarified.

Mr. VIVIAN. As I said, Dr. Hornig, it seems to me the essential authority contained in this bill which doesn't exist is that which lies in these back paragraphs and this particular authority I would not consider authorizing. I can understand that there may be reasons for not recopying the data with the mark under some circumstances or recopying it without the mark under some circumstances. However, it is not obvious to me why there is any objection to having it copied, except to control the profits being derived from it, and I see no reason to control the profits being derived from it.

Dr. HORNIG. I think this is tied to the question of user charges. If they develop it in any given form

Mr. VIVIAN. Yes.

Dr. HORNIG. The special effort and special expenses are going to be that for which you charge, then as I said I don't think you would prohibit the copying in the sense of use. But an independent republication would bypass the concept of user charges. So the abandonment of one

Mr. VIVIAN. I can explain why I object to some user paying a user charge. If a person buys this from a commercial firm, he will certainly pay the commercial firm's reproduction cost. If he were to do it on his own copying machines, it costs him money to run the copying machines. It seems to me there should be no objection that this data be used by anybody.

Dr. HORNIG. Clearly the idea is to disseminate data as widely as possible. But I think the problem of user charges comes up as I said when one starts thinking of more sophisticated data dissemination methods than a simple book or where aside from just the general compilations one has to go to special efforts to collect or produce data for special purposes. In such cases it seems to me quite proper that the user should bear some part of the cost of providing the services. Mr. VIVIAN. Could you give an example of that?

Dr. HORNIG. Yes, for instance, if I am going to set up various kinds of what amount to-I have no idea whether the Bureau in this connection intends to do these things-a retail distribution service of various sorts for the Government, that is, set up a distribution service where

in addition to producing the data, I send out various kinds of abstracts or indices of what is available to people. All of this costs me money and so do replies to requests for the data. I might set up a very elaborate machine for distributing the papers and pamphlets all over the country, and in the absence of any charges I can well imagine that I might get a lot of requests for thousands of copies of everything from many people who unlike me like to have their desks covered with piles of papers.

In this case I would consider this a waste of public funds, and I think one of the simplest restraints on this would be to have user charges.

Mr. VIVIAN. I disagree with your comments. It seems to me if the Government charges some fee for the use of the data, then very few people are going to acquire this data without expending that fee or some equivalent fee such as to a publishing firm or such as their own documentary reproduction offices. I think the purpose is not only to prevent the waste of printed matter, but also to inhibit its availability.

Dr. HORNIG. Well

Mr. VIVIAN. For example, the American Petroleum Institute may desire a certain compilation of data which nobody else wants and they are willing to pay a portion of the cost of gathering that datathen there might be some justification for user charges. But I can see no other justification for it.

Dr. HORNIG. I have no idea what situations in general might develop. I can see no justification for user charges on this kind of general provision of data.

But as I said, it seems to me that in the cases where special services or where special efforts must be made to collect the data, and this is the case you have cited, that it would. So, as a general matter, I think I would concur with you. Our effort is to get the widest possible dissemination of data. But the bill does not provide for the general use or the general restrictions. It provides authority to restrict under some circumstances, if I read it correctly.

Mr. VIVIAN. Thank you.

Mr. Chairman, I would appreciate it if the staff could review this I believe a far more lucid or less restraining phraseology is required, in section 7, in this regard.

Mr. DADDARIO. Mr. Vivian, the language in this bill is not so firmly fixed that it may not be changed as a result of these hearings. As we get into this question with other witnesses, we can be more precise than we can be today.

I believe that Dr. Hornig has given us the general philosophy concerning this bill without necessarily being specifically in support of this language.

Dr. HORNIG. Mr. Chairman, if I may add, as I said originally, the word "copy" in 7(b) does disturb me. It is republishing that one is really talking about, or—yes, it is commercial publication, I think. Mr. CONABLE. The question of enforcibility of this section I think is a very serious one, to try to prove copying of facts that may have independent existence elsewhere.

Dr. HORNIG. That is correct, although the commercial publication of compilations of data is enforceable.

Mr. DADDARIO. The discussion you have had with Mr. Vivian is going to be helpful to us because it points up that this is a problem. Mr. Vivian, we will follow it more carefully as we go along and if there are others who are as disturbed as you are, we will see what can be done about it.

Mr. VIVIAN. Mr. Chairman, I have just one or two other questions. Mr. DADDARIO. Yes, Mr. Vivian.

Mr. VIVIAN. I make these comments with reference to that section because, as I say, it is one of the principal new authorities conveyed to the National Bureau of Standards by the bill. Another is section. 8, which is effectively a punishment section.

Another new authority contained in this bill is in section 5 which gives Commerce the authority not to use the Government Printing Office. I understand that section is for the purpose of being able to fund ongoing programs with the profits derived from the sale of standard reference data as a part of the user charges concept. Here again, it is not obvious to me that it is in the best interest of the United States as opposed to the interest of the Department of Commerce that this money should not return to the Treasury.

Mr. DADDARIO. Dr. Hornig, you can answer if you like. This seems to me to be a question which we should go at great length into when Dr. Hollomon appears before us.

Dr. HORNIG. Mr. Chairman, I would prefer if you did that, because this particular point I have not carefully thought through. Mr. DADDARIO. Yes, we will do it that way.

*

Mr. VIVIAN. The next point relates to section 5 also, the words "person or agency" are listed there, on page 3, line 12: "Data ** may be published by a person or agency designated***" What does that mean? Is this another question you would prefer to hold for Dr. Hollomon?

Dr. HORNIG. I think I would prefer that he take that up.

Mr. DADDARIO. Yes, I would expect that we should probe more deeply into the terminology of all of these sections you refer to, Mr. Vivian, when Dr. Hollomon comes here. This is a matter of interpretation which he is best suited to answer.

Mr. VIVIAN. That completes the questions I had.

Mr. DADDARIO. Mr. Conable?

Mr. CONABLE. No more questions.

Mr. DADDARIO. Mr. Felton.

Mr. FELTON. I had one, Doctor. The whole objective of this program is that the raw data be evaluated by the best people in each field. I would assume that to analyze the data in any particular field would take a considerable amount of time. Is it realistic to ask an expert in a particular field to, say, take 6 months or a year off from the research he may otherwise be doing to perform this function for the Bureau?

Dr. HORNIG. Well, I don't believe it occurs quite that way. Dr. Astin can speak to this in more detail. But much of this evaluation function for any new numbers has to be performed by the people working in the field all the time, without taking any time off. On the other hand, it is realistic surely to have some people who are concerned with what is more mechanical, the compilation and the com