Lapas attēli

(a) When do you expect the National Standard Reference Data System to become substantially a computerized system? All of the above identified computerized portions of the National Standard Reference Data System are presently being studied or evaluated. In current NBS planning, fuil utilization of computers in most of these specific modes is two to five years away. Systemization of a complete computerized data file is the most difficult part of this system to plan for exactly at the present time. However, NBS computer scientists have estimated that all of the necessary computer components are either now in existence or will be fully developed by the time we are ready to utilize them. A computerized data file is probably at least five years away, and this minimum figure would be dependent upon rapid development of the critical evaluation programs to provide the data input to such a file.



1. How is it possible for the National Bureau of Standards to undertake the functions envisioned in this bill without significantly increasing the amount of research it performs in-house?

While some increase will be necessary in the number of people at the National Bureau of Standards who will be doing compilation work, as indicated by Dr. Hollomon in his answer to your question No. 7, no overall increase of any consequence in the amount of experimental research is anticipated. This is because the NSRDS program is concerned primarily with the compilation and evaluation of data that have been or are being produced by existing research projects in NBS and, most particularly, in other institutions. It is expected, of course, that the results of the evaluations and compilations will constitute an important source of guidance to the research program of NBS.

(a) Will it not be necessary for the Bureau to duplicate some of the experiments in-house in order to at least spot check the accuracy of the

various assigned values? Some spot checking of the various assigned values will be necessary. The Bureau will perform such checking only in those areas where it already has a high level of technical competence, and thus little additional research effort or research equipment will be needed for this purpose. In other technical areas, the spot checking will be done at other locations by the people who are doing the compilation. This is one of the reasons why we feel that compilation work should be performed by those who are expert in the field, and are actively working in it.

(6) After this program becomes fully operational, what percentage of the $20 million estimated annual cost do you expect to be performed in-house

and what percentage do you expect to be performed out-of-house? Table II attached to Dr. Hollomon's answers shows that on a dollar basis, three-quarters of the compilation work will be performed out-of-house and only one-quarter in-house. Providing the information services described on page four of my statement will be a major in-house effort and will require about 10% of the total estimated annual cost. Of the total program costs, approximately 40% will be spent for in-house activities and 60% for out-of-house.

2. On page 4 of your prepared statement you say that this storehouse will eventually contain a complete collection of compilations of critically evaluated data produced throughout the world.How many years after the enactment of this bill do you mean by "eventually?

While it is difficult to predict how long it would take for a comprehensive collection, it will probably take a minimum of three to four years to obtain substantial results with the exercise of proper precautions for high quality and careful coordination of the various projects. There will be some lag (averaging one to two years) between the time a project is started and the time its output is ready to be entered into the storehouse.

3. On page of your prepared statement you say that "hundreds of millions of dollars will be unnecessarily spent in the achievement of those goals unless NSRDS is effectively implemented.

(a) Can you be more specific as to what you mean by "hundreds of millions," and is this an overall or an annual savings?

(6) How did you arrive at that estimate? We estimate that savings of at least $100 to $200 million per year would result, and the figure may be considerably higher though this is difficult to estimate quantitatively. The figures quoted are based on the following analysis :

Each technical man is saved 10–12 minutes per week by having a convenient source of evaluated data more readily available. This adds up to 10 hours per year per man. Taking an average figure of $10/per hour per person (including overhead), this is a saving of $100 per technical person per year. For all of the


nation's 1.5 million scientists, engineers and technicians, the calculation gives $150,000,000 annual economy in time saved alone. Substantial additional savings would accrue through avoiding of errors incurred by the use of unevaluated data.

Other approaches to estimating the savings that can be expected were outlined in Dr. Hollomon's initial statement to the Committee on June 29, 1966.

4. On page 8 of your prepared statement you say that the narrower goals of VSRDS "is probably within the capability of available manpower in the United States.This being the case, where would the Bureau get the manporter for the broader goals which you describe?

The broader goal described would extend the range of the system to a much wider scope of materials and properties. We would recommend against such expansion in the immediate effort. It may become necessary to consider such expansion eventually, and manpower for the work would come from two sources: (a) growth of the technically trained manpower resources of the U.S. and (b) development of an international program which would utilize competent scientists all over the world.

(Whereupon, at 12:30 p.m., the subcommittee was adjourned to reconvene at 10a.m., Thursday, June 30, 1966.)






Washington, D.C. The subcommittee met, pursuant to adjournment, at 10:25 a.m., in room 2325, Rayburn House Office Building, Hon. Emilio Q. Daddario (chairman of the subcommittee) presiding.

Mr. DADDARIO. The meeting will come to order.

Our first witness this morning is Dr. Frederick Seitz, President, National Academy of Sciences. Dr. Seitz, we want to welcome you here this morning. We are pleased to have you before us again.



Dr. Seitz. Thank you, Mr. Chairman; it is always a pleasure to appear before your committee.

I have a prepared text. However, I will deviate from it at times in view of some of the previous testimony.

The National Academy of Sciences is thoroughly in accord with the objectives of the Standard Reference Data Program at the National Bureau of Standards. We feel strongly that the program should be expanded and strengthened since further advances in science and technology will depend heavily upon the availability of reliable quantitative scientific information, particularly that which can be identified as critical tables of standard reference data.

Certainly the most notable example of an effort to provide critical tables of standard reference data is the International Critical Tables of Numerical Data of Physics, Chemistry, and Technology. Preparation of these tables resulted from discussions begun at the 1919 Conferences of the International Union of Pure and Applied Chemistry at London. The United States was assigned financial and editorial responsibility for the project and the National Academy of SciencesNational Research Council accepted the executive, editorial, and financial responsibilities for the United States.

The entire enterprise was made possible by the cooperation of the American Chemical Society and the American Physical Society, together with essential support from industry, which contributed funds totaling $200,000. This famous collection of numerical data was the result of cooperative efforts by some 400 scientists in 18 differ

« iepriekšējāTurpināt »