Lapas attēli

decision will be shown on the annotated list by the use of a decision code which will be included in the instructions provided by the General Services Administration.

ROBERT P. MAYO, Director.

Chairman PROXMIRE. I understand, Mr. Staats, that you will give us a 15 to 20 minute review of your prepared statement which will be included in full in the record. We are delighted to have you.


Mr. STAATS. Thank you very much, Mr. Chairman.

I will brief my statement and try to hold within 15 minutes, recognizing you have other witnesses here this morning.

In the prepared statement, we bring out the growth in the Government's investment in computers, as you have already pointed to in your opening remarks.

Next, we have simply here listed a number of reports that we have done in this field to give you some notion as to how active we have been in our interest in this subject.


Next, we point out the early work of the GAO in recommeding a better inventory of automatic data processing equipment, which had something to do with the Budget Bureau Circular which was issued in 1959, and which has led to the annual publication of inventory.

The most recent Budget Bureau issuance on this is referred to in the prepared statement, and they will, I am sure, go into this in greater detail: Circular A-83, which prescribed a management information system for governmentwide use.

Now, although the executive agencies have been and are now required to submit information on their computer resources in accordance with BOB Circulars A-55 and A-83, our reviews have shown that the reporting system does not necessarily produce the accurate, complete, and useful information that is necessary to facilitate the making of proper management decisions on procurement and the utilization of APD resources.


First, as we point out in the prepared statement, there is need for realistic and timely projections of acquisitions and releases of ADP equipment by the Federal agencies to improve reutilization efforts to provide assistance for use in Government-wide contract negotiations and also to prevent unneeded ADP purchases.

There is also a need for inclusion of information regarding software and its use in Government operations-to reduce duplication of effort and unnecessary costs.


We were told by some Federal agencies that little use had been made of the management information system because the system lacked current and reliable information. For example, computer printouts of the June 30, 1969, reports were not available to GSA until December 15, 1969, and distribution of copies of these reports to the agencies was not made until February 20, 1970. So there is a considerable lag here. We don't minimize the difficulty, Mr. Chairman, of maintaining an accurate, complete and timely inventory, but we do feel that it is important to point out that the present inventory is not having the full intended effect because of these difficulties.

With respect to the peripheral equipment acquisition from independent peripheral manufacturers that you referred to, we covered that in the prepared statement. We made an early study of this subject, I might point out back in April of 1968, and then following the hearings and the result of our study, as we point out in the prepared statement, we went into this subject in greater detail. We conducted a separate study, issued our report to which you have referred, the Study of Acquisition of Peripheral Equipment for Use with Automatic Data Processing Systems, which was issued June 24, 1969. (Text in app. II.)


The study pointed out that it is common practice for Government ADP managers to obtain all required ADP equipment from computer systems manufacturers even though certain items of equipment can be procured more economically from the original manufacturers or from alternate sources of supply.

We identified selected computer components that are directly interchangeable (plug-to-plug compatible) with certain other systems manufacturers' components and are available at substantial savings. We found that a number of private organizations had installed available equipment of this type and had achieved substantial savings. Yet we found only a few instances where Federal agencies had availed themselves of this economical means of acquiring computer components. We expressed the belief that central agency leadership could provide impetus for achieving similar savings in the Federal Government.


We estimated that, if plug-to-plug compatible components were rented from independent manufacturers rather than from systems manufacturers, annual savings would amount to at least $5 million. We estimated also that, if such components were to be purchased, they could be purchased for $23 million less from the component manufacturers than from the systems manufacturers.

We also expressed the belief that, in addition to the estimated savings in acquiring plug-to-plug compatible components, savings are available in the acquisition of non-plug-to-plug components from sources other than the systems manufacturers. We estimated that the purchase cost of such components-now being leased for about $50 million a year-from the systems manufacturers would be about $250

[ocr errors]

million whereas the acquisition price for similar components from an alternative source of supply probably would be about $150 million— a difference of about $100 million.


One of the problems associated with the use of non-plug-to-plug components involves the compatibility of components with the main computer system. In this regard, the state of the computer industry today is such that, with the exception of plug-to-plug compatible peripheral devices, components cannot generally be directly interconnected with other manufacturers' components or systems. In this respect, both an electronic and a software interface generally have to be provided before the equipment can be interconnected.


A solution to this problem, which is now being considered by the industry, is the possibility of standardizing the interface media between peripheral equipment and the central processing unit. Interface standardization would stimulate competition in the peripheral equipment industry and would allow the user to select the peripheral equipment best suited to its requirements.

To this end, the American National Standards Institute, a privately supported organization acting as the national clearinghouse and coordinating agency for voluntary standards in the United States, has created a committee to consider the feasibility and practicality of input/output interface standardization.


Although the committee has been in operation since early 1967, progress has been slow in accomplishing desired objectives.

We believe that the development of a standard interface will promote industry competition and result in certain economies. It will provide the users with increased flexibility in the selection and use, regardless of the manufacturer, of those components best suited to achieve the desired objectives. Under such circumstances, the users will be in a better position to match system specifications with available equipment.


It is our view that, if an industrywide standard cannot be established in the near future, the National Bureau of Standards should be directed to develop a Federal standard interface program in order to achieve the significant savings which should result from increasing the compatibility of major components with main frame equipment.


We have been advised that the Bureau of Standards has been handicapped by a shortage of funds in this area, and we believe, Mr. Chairman, that this is a matter that not only this committee but other committees of Congress should particularly consider because this is really

the key to what we are talking about in our report. Unless greater effort on the Government side is devoted to this, we do not think that we are going to be able to make very satisfactory progress.

Now, the Bureau of Standards currently has some 20 standards in process, as I understand it. They have only 14 professional staff members to do all of this work. Only one-half of 1 man-year is estimated to be devoted to this area of interface which is the key to the purchase of the compatible equipment that we are referring to in our report. We believe that this is a matter that you may wish to go into with the Bureau of Standards, but we certainly believe that this is not in proportion to the importance of the problem.

I might say I have read this part of my report because I think it is the main story that we have to set forth to the committee this morning.

I will skip over the page in the prepared statement where I point out that the Budget Bureau has issued a bulletin on the acquisition of peripheral components for installed ADP systems. I am sure they will go into this with you in great detail.


We also have in process currently, as a GAO study, a follow-up on our report, to see what steps have been taken to implement recommendations in our report.

The lead on this study is being taken by our Boston office working with our other regional managers, but I must say that progress in acquiring peripherals from independent component manufacturers to date is really pretty discouraging. I will cite just a couple of instances and then I will be glad to insert more information on our study in the record if you wish.

For example, as of April 6, 1970, the Army had issued an RFP for only four plug-to-plug compatible tape drives. The Navy has completed one procurement for 40 tape drives; a procurement is in process for about 20 tape drives; and a third procurement is being prepared for a Navy and Marine Corps replacement program that will involve about 500 tape drives.

The Air Force has identified about 155 second generation tape drives which can be replaced by plug-to-plug compatibles. However, these tape drives are being used on computer systems that are scheduled to be phased out and have an average remaining system life of less than 10 months. We have many examples of this and we will be glad to put these in the record if you wish. But I did want you to know we are following up on our report.

Chairman PROXMIRE. Yes, without objection that will be put in the record.

(The information referred to follows:)


To: Deputy Director, OPSS-E. J. Mahoney.

From: Regional Manager, Boston-Joseph Elder.

Subject: Follow-on to the Study of the Acquisition of Peripheral Equipment for

use with Automatic Data Processing Systems.

Ever since the subject report was issued we have been following up on the actions Federal agencies were taking as a result of our recommendations or suggestions. In addition to the favorable actions summarized by your Office

in March 1970, there was the letter received from the Veterans Administration and the General Services Administration (GSA) letter concerning its "actions" on plug-to-plug replacements.

We all know that it is far easier to issue regulations, policies and requirements than to police these requirements. Studies such as described in Department of Defense's letter of August 21, 1969 to the Comptroller General in response to our report and permissive regulations such as issued by GSA often are followed by further delays in implementation or complete inaction.

It is now a year since the report was issued. In view of the significant sums involved we believe that we should now find out and report on the actions actually being taken on our recommendations. We should examine into the delays, and attendant losses in those instances where no action was taken which should have been taken. Additionally we should point out the need for more effective management of the vast Government ADP resources. Our plan for this follow-up review is enclosed. To date we are concerned with the first priority of the enclosure. Our reasons for this are due to the information we have gathered in following up on this matter, as summarized below.

We have contacted manufacturers and distributors of plug-to-plug compatible peripheral equipment to determine the extent of their sales and rentals to Federal agencies. Of 18 firms contacted 5 have had sales or rentals to the Federal Government, 11 had no Federal sales or rentals when contacted and returns from two vendors have not yet been received. The total quantity of installed plug-toplug compatible tape drives and disk drives sold to date is 229; associated firm savings are about $1.3 million.

In a contact with Mr. L. R. Caveney of Bryant Computer Products he expressed dissatisfaction with GSA negotiations for a Federal Supply Schedule. He stated that GSA insists on treating his company like a total systems manufacturer and refuses to recognize that use of his disk drives simply requires switching plugs.

One of the vendors contacted expressed concern over plans GSA, the Army, Navy or Air Force may have to issue large quantity RFP's for plug-to-plug compatibles. He explained that very large quantity buys would severely limit the number of vendors who could be responsive and that an economic order quantity for procurements should be used to insure competition among a number of vendors.

As of April 16, 1970 the Army had issued an RFP for only four plug-to-plug compatible tape drives. The Navy has completed one procurement for 40 tape drives; a procurement is in process for about 20 tape drives; and a third procurement is being prepared for a Navy and Marine Corps replacement program that will involve about 500 tape drives.

The Air Force has identified about 155 2nd generation tape drives which can be replaced by plug-to-plug compatibles. However, these tape drives are being used on computer systems that are scheduled to be phased-out and have an average remaining system life of less than 10 months. Under a delegation of authority from GSA, the Air Force has released a letter of intent to vendors for replacing these tape drives with rented plug-to-plug compatibles. Under these conditions it may be difficult to obtain peripheral vendors interested in installing and maintaining a piece of equipment for such a short period of time. In fact, many of the systems may be phased out before the planned conversion to plug-to-plug peripherals can take place. Since GSA is currently undertaking a Government-wide replacement program (see item below) the question arises as to why the replacements being planned by the Air Force could not be purchased rather than rented and then be transferred to other Federal installations when their use is no longer required by the Air Force.

The Air Force Directorate of Data Automation is currently working on another list of peripherals for third generation computer systems for a replacement program with plug-to-plug compatibles. However, the Air Force policy on new system procurements continues to be to procure entire systems from systems manufacturers.

Utilizing data contained in the ADP/MIS, GSA has identified 2,867 rented peripheral components which if replaced by rented plug-to-plug compatibles would reduce annual rental costs $8.6 million; if all items are purchased a $31.4 million cost reduction is possible.

On February 16, 1970 GSA issued each Federal agency a listing of those peripherals rented by that agency which could be replaced by plug-to-plug compatibles. Each agency was requested to endorse this listing and indicate which

« iepriekšējāTurpināt »