NSAM 368 (INTELLIGENCE INFORMATION-HANDLING SYSTEM)

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP82M00097R002600090004-1
• Release Decision: RIPPUB
• Original Classification: C
• Document Page Count: 43
• Document Creation Date: December 22, 2016
• Document Release Date: November 13, 2009
• Sequence Number: 4
• Publication Date: May 14, 1968
• Content Type: MF

Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
%_4 Vl ? 1 A. Lill \ J. 1 i JLL _
USIB-D-71. 1/2
14 May 1968
Limited Distribution
UNITED STATES INTELLIGENCE BOAR D
MEMORANDUM FOR THE UNITED STATES INTELLIGENCE BOARD
SUBJECT : NSAM 368 (Intelligence Information-Handling
System)
REFERENCE : USIB-D-71. 1/1, 22 April 1968, Limited
Distribution
1. The attached copy of a memorandum, on the subject for the
Director of Central Intelligence (DCI) from the Special Assistant to
the President for National Security Affairs with respect to the
response to NSAM 368 by the DCI enclosed with the reference paper,
is circulated for the information of the United States Intelligence
Board (USIB).
2. Mr. Helms has referred the attached memorandum from
Mr. Rostow to Mr. -Bross., D/DCI/NIPE, for appropriate action in
preparing, with the assistance of all USIB member agencies, a
response to the request for a further report by October 22, 1968.
U
NSC review
completed. Attachment
NSA review
completed
DIA review(s)
completed.
State Dept. review
completed
Executive Secr a
CONFIDENTIAL
GROUP 1
Excluded from automatic
downgrading and
declassification
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
DENTIA'
WASHINGTON
May 8, 1968
R.ol.u~r
MEMORANDUM FOR THE DIRECTOR OF CENTRAL INTELLIGENCE
SUBJECT: NSAM 368 (Intelligence Information-Handling System)
Please refer to your memorandum for the President dated
April 22, 1968, in response to the provisions of NSAM 368
which called for (1) the preparation, in consultation with
the President's Foreign Intelligence Advisory Board, of a
plan for the phased implementation of a community-wide infor-
mation-handling system for the secure and efficient processing,
storage, retrieval and dissemination of intelligence information,
(2) a discussion of alternative ways to manage such a system,
together with your recommendation with regard thereto, (3) an
identification of the costs and benefits of each phase of the
program including a detailed proposal for the first annual in-
crement as part of the FY 1970 budget, and (4) submission of
the plan, with the concurrence or comments of the Departments
of State and Defense, to my office by May 1, 1968.
Your memorandum of April 22 provides useful information as
to the progress made thus far in planning for improved information-
handling systems within the U. S. intelligence agencies, and in
identifying the preliminary actions to be taken in preparation of
a long range plan looking toward the development of a community-
wide system. While it is noted in your memorandum that you are
presently unable to indicate the time phasing of actions that
remain to be taken, it is hoped that your next submission in
response to the NSAM can provide the President with a more
definitive picture of the projected community-wide system. It
would also be helpful to have a report on the progress made-by
the individual departments and agencies in the development of
information-handling systems which meet their respective needs
and fulfill the requirements of the projected over-all system;
the estimated costs allocated to successive phases of the
development of the community-wide system; and a more precise
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
N..
estimate as to time phasing. Accordingly, it is requested
that there be furnished to this office and to the President's
Board by October 22, 1968, a further report along these lines,
including information responsive to the requests set forth in
NSAM 368 as enumerated in the first paragraph of this memorandum.
W. W. Rostow
cc: The Secretary of State
The Secretary of Defense
Director, Bureau of the Budget
Special Assistant for Science and Technology
Chairman, President's Foreign Intelligence Advisory Board
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
CONFIDENTIAL (When Filledln )
LIMITED DISTRIBUTION LIST
Intelligence Board Members and.
Service Intelligence Chiefs
No. Copies
State
DIA -2 Additional for O/SecDef
NSA -6 Additional
AEC
FBI
Army
Navy
Air Force
USIB-D- 71.1/2
Date 14 May 8
Classification CONFIDENTIAL
USIB Committees No. Copies
CCPC
C-pp1B- Secretary IHC -1
COMIREX
EI C
GMAIC
JAEIC
SECURITY
SIC
SIGINT
WATCH
CIA Internal
DCI
DDCI
Executive Director
D /DCI /NIPS
DD/I
DD/P
DD/S
DD/S&T
D/OCI
D/ONE
D/OSI
D/OER
D/CR S
D/BGI
D/OSR
D/DCS
C /IRS
Director, NPIC
Director, Security
Director, Communications
General Counsel
IG
Chief, Estimates Staff, ONE
ER/Vital Records
USIB Secretariat -20
Executive Secretary, NSC
President's Foreign
Intelligence Advisory Board
Special Instructions
CONFIDENTIAL (When Filled In)
NIPE Staff 25X1
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 x,....o*" Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
Copy No. 5 9 of 80
USIB-D-71. 1/1
22 April 1968
Limited Distribution
UNITED STATES INTELLIGENCE BOAR D
MEMORANDUM FOR THE UNITED STATES INTELLIGENCE BOARD
SUBJECT Response to NSAM 368 by the Director of
Central Intelligence
REFERENCES a. USIB-D-39. 1/8, 29 March 1968
b. Memorandum for Holders of USIB-D-39. 1/8,
1 April 1968
c. USIB-M-503, 4 April 1968, Items 4 and 5
d. USIB-M-504, 11 April 1968, Item 6
1. The enclosed copy of a memorandum for the President from the
Director of Central Intelligence (DCI) in response to NSAM 368 is circulated
herewith for the information of the United States Intelligence Board.
Z. The enclosed memorandum reflects the Board's discussion and
actions in references c. and d. It also incorporates as Annexes E - H
of Attachment A the summary reports submitted by the Department of
State, CIA, DIA and NSA as requested in reference b. A single copy of
the detailed reports, which are to be forwarded separately as Attachment
B to the Memorandum for the President, will be held in the Intelligence
Information Handling Committee (IHC) Support Staff for reference.
d
Enclosures:
Distribution List Attached
Discussed at 11 Apr 68 USIB
Meeting, USIB-M-504
See IHC Comm File for
copy as presented to Pres.
Executive ecre y
SECRET
GROUP 1
Excluded from automatic
downgrading end
declassification
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
N L V 1% L 1
Aftv~
IL 1"e
THE DIRECTOR OF CENTRAL INTELLIGENCE
WASHINGTON, D. C. 20505
22 April 1968
SUBJECT: NSAM 368
This report is submitted in response to the instructions
contained in NSAM 368. It covers projects now under way in the
intelligence community which will contribute to the development
of a system to improve the dissemination, processing, storage
and retrieval of intelligence information throughout the intelligence
community. Specifics are appended as Annex A to provide an outline
of what must be done to arrive at a long range plan for improved
information handling. These include the categories of studies and
experiments that will be necessary, together with indication of the
time phasing of their relationships to one another. It will take
some time to get this all done. I can't tell yet just how long.
A community-wide information handling system will follow
from the continuous improvement and integration of the component
agency systems, which are being automated as rapidly as feasible.
Some of the automated agency systems are already fully operational
and others soon will be. They will serve to expedite and facilitate
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
the exchange and availability of information among agencies and
contribute to the elimination of undesirable duplication in filing and
processing information.
Subsequent to the issuance of NSAM 368, USIB approved a
Statement of Objectives for Intelligence Information Handling. The
Statement provides guidance for the continuing activities which
must be undertaken to ensure the orderly development of a
community-wide system for intelligence information handling.
USIB has also approved a revised DCID 1/4, establishing a new
Intelligence Information Handling Committee with terms of reference
tied directly to the Statement of Objectives. These two actions
(Annex B) establish the policy of USIB concerning the intelligence
information handling problem.
Integration of intelligence community efforts in the field of
information handling will proceed under my direction. With the
advice of USIB and using the Chairman of the newly approved
Intelligence Information Handling Committee as my direct representa-
tive, I expect to achieve effective management control. I believe that
these arrangements will ensure that the experience and talent of
the USIB member agencies are effectively pooled and at the same
time will preserve the developmental and support facilities required
to meet the specialized needs of individual agencies.
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
VWVI SECRET *MW
An inventory has been made with the help of the member
agencies of those activities and systems that relate to intelligence
information handling. Attached are a status report on the Community
On-Line Intelligence System (COINS) (Annex C), a listing of other
selected community projects (Annex D) and summaries of activity
at State, CIA, DIA and NSA (Annex E through H). A more detailed
report covering these four intelligence agencies is being forwarded
under separate cover as an additional attachment to this report.
Accomplishment of a community system will require a number
of adjustments. While these will not include alteration in the basic
responsibilities of agencies, they will involve financial arrangements
and coordination of data exchange to insure that the financial burden
does not fall disproportionately on one or a few agencies. I will take
these up with the appropriate managers as the occasion requires.
It is clear from the work already accomplished that the
development of any information handling system on an agency or
community basis requires allocation of resources well beyond those
made available in the past. This has not been a seriously inhibiting
factor to date because much of the work so far has been of an
experimental and developmental nature. We are now in the equipment
installation and operation stage in many of the agencies and the
3
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 ? Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
'4W SECRET NUIP,
requirement for resources will mount rapidly. This will affect
budgets throughout the community.
The significance of these points, and others, will become
clearer as the detailed planning proceeds. I will report to you
as significant milestones in the program have been reached.
Attachment A: Annex A through H herewith
Attachment B: Forwarded under separate cover
4
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
wall
C-O-N-F -I-D-E-N-T -A-L
ATTACHMENT A
Annex A - Outline of Some Preliminary Action to be Taken
in Preparation of Long Range Plan.
Annex B - USIB Statement of Objectives for Information Handling
and DCID No. 1/4 Establishing a New Intelligence
Information Handling Committee.
Annex C - COINS Summary Report.
Annex D - Selected Community Information Handling Programs.
Annex E - State Department Summary.
Annex F - CIA Summary.
Annex G - DIA Summary.
Annex H - NSA Summary.
C-O-N-F-I-D-E-N-T-A-L
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
C-O-N-F-I-D-E-N-T-I-A-L?'r
OUTLINE OF SOME PRELIMINARY ACTIONS TO BE TAKEN
IN PREPARATION OF LONG RANGE PLAN
PHASE I
Studies .
Definition of the information handling problem.
Analysts' processes and uses of information.
User requirements for information service.
Future communications requirements for information
exchange.
ADP security problems.
Rapid input systems.
Rapid search and readout systems.
New concepts for exchange and interaction.
Document storage and retrieval by use of videofile,
closed circuit television, LDX or other means.
Relative costs of file duplication versus communica-
tions costs to achieve sharing of files.
Experiments.
COINS.
Tape exchange.
Overnight information updating via COINS.
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
C-O-N-F-I-D-E-N-T-I-A-"e
Experiments (con't).
Secure circuit voice response.
Exchange of formatted reports of end products
by various means.
Exchange of microform-photochip.
Evaluation.
Definition of Phase II and plan for next generation
of community information handling system.
Implement Phase II plan for community information handling
system.
Expand use of system.
Improve timeliness of response.
Incorporate interactive problem-solving into system.
Incorporate graphic operations and image exchange into
system.
Further improvement of Phase II as state of the art permits.
2
C-O-N-F-I-D-E-N-T-I-A-L
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
Annex B as forwarded consisted of the following
USIB-approved attachments to USIB-D-71. 2/ 1:
A - USIB "Objectives for Intelligence
Information Handling"
B - Director of Central Intelligence
Directive (DCID) No. 1/4,
"Intelligence Information Handling
Committee"
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
1 CONFIDENTIAL
Community On-Line Intelligence System (COINS)
1. Background. This is a brief progress report on the COINS
experiment which is being conducted under the general supervision of DIRNSA
in order to implement the President's Foreign Intelligence Advisory Board
Recommendation No. 2 of 15 June 1965. The concept calls for DIA, NSA
and CIA to maintain COINS files in their own information retrieval computer
systems which will be connected by secure data links through a central
switch at DIA. The Department of State and the National Indications Center
(NIC) will have remote query consoles connected to the DIA computer through
which they can interrogate COINS files.
2. Significant Milestones and Current Status
a. On 21 December 1966 a secure 300-character-per-second
(2400 BPS) data link was installed between an NSA computer complex
and the DIA network switch.
b. The first test message was transmitted on 13 February 1967.
The first successful interagency test between a remote terminal at NSA
and the computer at DIA and between a remote terminal at DIA and the
computer at NSA occurred on 20 April 1967 using unclassified files.
c. In early September 1967 NSA and DIA selected COINS files
were made available and the COINS links between NSA and DIA was activated
C.UNFIDENTIAL
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
1W` CONFIDENTIAL
for two hours a day, two days each week. On 2 October 1967 the link
became active for two hours each day, five days a week, and on
12 January 1968 for six hours per day, five days per week.
d. The system is currently undergoing further testing to detect
and eliminate system faults. A date when the NSA-DIA link can be declared
"fully operational" twenty-four hours a day, seven days a week, cannot
be stated at this time.
e. Circuits from the central switch to CIA are being abtivated
at the present time. The Department of State and the National Indications
Center will be connected by July 1968.
f. On 11 April 1968 the DCI designated DIRNSA as executive
agent to carry out the balance of the experiment. The DCI will provide
guidance to DIRNSA through the NIPE staff.
3. Training and Familiarization. The COINS experiment still
requires a concerted effort to acquaint the user analyst with the system
files and the procedures required to access them. Although the COINS
system has been kept simple and small, minimal instruction to permit
the analyst to interact with these files requires careful preparation and
presentation.
C.ONFIDENTIAT,
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
V CONFIDENTIAL =..PF
4. Test and Evaluation , The testing of COINS will start in July
1968, with evaluation to follow. This process will run for one year with
a final report to be submitted to USIB prior to the end of December 1969.
CONFIDENTIAL
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 ? Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
Selected Community Information Handling Programs
1. Biographic Intelligence. Selected biographic files are
now available in the COINS experiment. An investigation of biographic
activity has led to the selection of these files and to the development
of new projects in the security name check system. This investigation
has been expanded to determine what biographic intelligence would be
useful to a community-wide information system and how it ought to be
handled in the system.
2. Research and Development. The desirability of undertaking
selected R&D projects and of exchanging information on them is
recognized and progress has been made, including the establishment
within CIA of an R&D experimental facility for the development of new
techniques in information processing. NSA has an established R&D
facility with information handling related projects under way, and
DIA is in the process of developing a capability in this area. As soon
as qualified personnel can be assigned, an R&D subcommittee of the
new Intelligence Information Handling Committee will be formed to
carry out the stated USIB objective in this area.
3. Training. The individual agency training programs in
information science have been expanded and during the past year
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
C-O-N-F-I-D-E-N-T-A-L
various courses running in length from three days to a full academic
year were attended by 8166* persons. Sixteen percent of these courses
were at universities or technical training schools. The total training
given amounted to approximately 350, 000 mar hours.
The following table summarizes the information science
training under four broad categories:
Type of Training
No. of Persons''`
%
General Orientation
2003
25
Methods and Techniques
2408
37
Computer Programming
3058
29
Systems Analysis, Design
697
9
and Implementation
816
100
As a service to the intelligence community, the Defense
Intelligence Agency established an Information Science Center in the
Defense Intelligence School in December 1967. The first regular
courses open to students from the intelligence community were
scheduled to start in January 1969. There has been considerable
difficulty in recruiting a competent faculty within the civilian grade
structure authorized, and this may delay the start of the courses.
*These figures do not accurately reflect the actual number of individuals
since some - perhaps as many as one half - took more than one course
during the past year.
2
C-O-N-F-I-D-E-N-T-A-L
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
C-O-N-F-I-D-E-N-T-A-L
4. USIB-approved Programs. USIB has also approved the following
programs which are now under development:
a. The Automated USIB File and Program
Catalog System;
b. The USIB Item Register;
c. The USIB Installation Register;
d. The Content Control Code;
e. Photo-chip Standardization.
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
vwp~ No
The Department of State Substantive Information stem
The Department of State has recently completed its analysis of present
substantive information handling practices and, based on the findings,
has designed a modern information system. It should be noted that the
terms "substantive information" (as used in the Department) and
"intelligence" (as used in the intelligence agencies) are generally
synonymous. The systems, both present and future, are described in
detail in "A Modern Information System for the Department of State",
December 1967. A brief synopsis follows below.
1. Present Practices
a. Dissemination: The present system of dissemination is characterized
by a dispersion of entry points for various types of information.
The Communications Center receives and distributes telegrams, the
Records Services Division receives and distributes Airgrams and
other pouch mail, and the Bureau of Intelligence and Research
handles all input from intelligence components of other agencies.
In addition, a large volume of substantive information comes
into the Department addressed to individual bureaus and is handled
by the addressee. All documents not directly addressed to individuals
or components are read and distributed in terms of requirements
submitted (generally annually). These requirements are not the
end users' but reflect only the consolidated needs of the Bureaus.
Consequently, dissemination is made only to the Bureau level. Here,
a second reading of the documents is required to determine the needs
of offices within each Bureau, and in some cases two or more
additional readings are performed before a document reaches a desk.
Direct address mail goes to the addressee and is usually not
disseminated further, regardless of who might have need for the
information. No mechanism exists for alerting others to the
existence of such information.
The dissemination process in the Department is entirely manual.
It is confined to the distribution of documents and is oriented
toward the handling of paper rather than information.
b. Storage and Retrieval: Existing practices in the Department of State
for preserving foreign policy information and enabling officials to
use such information consist of storing documents in files and using
file folder headings to locate documents containing needed information.
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
Nov low
ANNEX E
There are three types of files in general use in the Department-
Central Foreign Policy files, authorized Decentralized files, and
individual working files. Since the establishment of these various
files is based on the principles of records management rather than
information management, it is impossible to find required infor-
mation on any problem in any one place. The Central Foreign Policy
File is the major repository for all Airgrams and telegrams, but
the 33 authorized decentralized files held by designated bureaus
for subjects of prime interest to them contain most other materials.
All intelligence agency produced materials, for example, are filed
in the Bureau of Intelligence and Research. Bureau working files
contain a variety of materials including communications, publications,
intelligence and other information from foreign and domestic sources.
Many documents are duplicates of those maintained in the Central
Foreign Policy File, the decentralized files, and other bureaus'
working files.
Files are organized by subject matter and generally follow the
Department's Records Classification Handbook. Most offices also
maintain a chronological file. As stated above, the Central
Foreign Policy Files is far from complete. All offices must
retire material which is three years old. In the five years
following this retirement, the material is screened for documents
which properly belong in the Central File. Since documents of
"continuing interest" may be excluded from retirement and retained
by an office, the central repository still remains incomplete even
after eight years.
To retrieve information, officers rely heavily on their own working
files. Occasionally, they will call on the Central Foreign Policy
File, or other files known to them. In most cases, however, if the
required information is not in their own files, they will contact
a colleague in the Department or in another agency who might throw
light on the problem or have a pertinent document.
In summary, the Department's files are set up for the preservation
of records rather than to facilitate the use of information. The
Department has no adequate retrieval system. As a result, the
individual officer overtaxes the various dissemination systems to
stay alert to relevant information, and he spends much time and
effort in building his own file. Despite this, he still has
incomplete information.
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
c. Collection Guidance: There are three principal ways in which
collection requirements are generated. First, the Foreign
Affairs Manual Series provides broad guidance in the fields
of political and economic reporting. Second, the Department
relies heavily on field initiative. The reporting officer
is expected to use his judgment to report anything he deems
significant "even though it is not listed in...outlines."
And, finally, the reporting officer responds to specific
queries from the Department. There is, however, no effective
coordinating organization to supervise the systematic collection
of needed information. The Foreign Reporting Staff of the Bureau
of Economic Affairs checks requests for economic/commercial infor-
mation against a set of agreed criteria before passing it to the
field for collection. Requests for Biographic information, maps,
and photographs are passed to another agency for search of an
existing data base to determine if the information is already
available. For all other categories, the user may levy his
requirements directly on the embassy he deems most competent to
collect the desired information. In the absence of a selective
dissemination and a reliable retrieval system, the latter route
is often the only means a user has available to get information.
2. Proposed Information System
The future system in the Department of State will be an integrated,
computer based, on-line document, information, and data system.
End users, rather than bureaus, will submit detailed dissemination
requirements to a central processing organization. User profiles
will be entered into the computer. All documents and information
(with a few exceptions) will enter the Department through this
central organization. Documents will be indexed and the index
matched against the profiles to determine dissemination. The user
will have the option of receiving documents (for subjects of prime
concern) or periodic proxies (accession lists, extracts,etc.) which
may serve his peripheral interests. He will be able to change his
profile daily, if necessary to adjust to changing requirements.
The same index records will also serve future recall. While the index
records reside in the computer, the document will be retained in some
form of microfilm. It is recognized that some special collections of
documents must remain outside the central system. The index records
to these collections, however, will be entered in the computer so
that all queries can be levied against a central, single reference
point. The recall sub-system will also be able to provide proxies,
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
ANNEX E
such as summaries and abstracts in place of the full document. In
addition, the system will contain automated information files, e.g.
political parties/elections, foreign scientific relations, treaties,
laws, agreements; and data files, e.g. country data, tariff data,
demographic statistics, gross national product. The system will
include information and data files of importance to the conduct of
foreign affairs created by or built jointly with other agencies.
The existence of a reliable recall and dissemination system will
make it possible to establish a collection guidance sub-system which
will continuously review the Department's needs, monitor collection
capabilities, maintain a requirements registry, and coordinate the
collection activities of the Department with those of other agencies.
The new system will provide the user with fewer documents and more
pertinent information. It will remove the need for multiple reading
of documents at various levels by reacting directly to the end users
needs. The central reference service will make the total data base
available to the user, and he will have data and information available
in addition to documents.
3. Existing Automated Systems
The Department has two major automated systems in operation. One is
the Automatic Terminal System (ATS), a computer controlled communications
switching system. The other is the External Research Publishing and
Retrieval System (XPARS) described in Attachment 3.
The Department (in Washington) currently uses approximately 400 man/
years to perform the document handling functions described above at
a cost of about $4,000,000 p/a. Without modernization, it is estimated
that this cost will increase to about $4.8 million in the next five
years. The proposed system is to be developed over a five year period,
will require.369 people and will be operated on an annual budget of
$4.8 million (including hardware). The development cost is estimated
at $5.5 million. It should be noted that no funds are available at
this date to proceed with the implementation.
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
`rr S-E-C-R-E-T Ifto,
ANNEX F
Central Intelligence Agency
I. Information Processing Program
1. The CIA. Information Processing and Exploitation Program
for FY 1969-73, which is now being processed as part of our annual
PPB cycle is as follows:
2. The fundamental objective of the Central Intelligence Agency's
information processing and exploitation program is to support the analyst
who produces finished intelligence and the operational personnel who
collect information and carry out operational assignments. Guided by
the intelligence requirements and operational decisions of policy officials,
analysts and operators need information to make informed judgments
and estimates. To make such judgments and estimates themselves
and to evaluate the work of others, producers of finished intelligence
must have access not only to the raw data but also to the methodology
by which the data may have been processed. Although such access is
within the computer state-of-the-art and is provided in specialized
automated and manual files, its incorporation into a community
computer network demands a sophistication of software design and
a computer security environment which are clearly not within the
current state-of-the-art. Both because of previous limitations in the
state-of-the-art and the absence of persuasive evidence of gain to
the analyst, the Agency has proceeded cautiously with the creation of
large automated central files. In an effort to control the input of
information to his files and the way the information is manipulated,
including the proper application of security compartmentation, the
analyst has preferred limited access files whether or not the files
are automated.
3. The large central storage and retrieval files of the CIA
(and other intelligence agencies) have been used mainly to meet
general reference requirements. They have tended to contain index
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
S-E-C-R-E-T NO,
ANNEX F
or pointer data to documents or to other files rather than to contain
information per se; a major goal of more recent system design has
been provision of more rapid and accurate information retrieval.
Several attempts in the USIB Community to create such files have
died a borning and others have been characterized by low use rates,
unsatisfactory response (in terms of analyst requirements), and,
high maintenance costs. We have renewed this effort in the light
of the latest advances in the technology of information handling but
we intend to proceed cautiously in view of the serious problems
remaining, particularly the high costs for input to such files.
4. Although we have sought to keep abreast of advanced
technology and to apply it as it proved feasible and desirable, we
have not installed automated methods for their own sake. In many
instances, we have found that manual methods provide the only
economical and workable solution to our data handling problems.
We have constantly improved these methods, and replaced them
as appropriate, with automated methods. Typically, however,
automated systems have involved more than the mere mechanization
of existing processes. Rather, they have involved the redesign
of files and procedures and the generation and manipulation of
data which could not have been handled by manual systems. Finally,
automated systems have themselves been constantly reviewed and
improved where such action was needed. We have kept in mind
that automated systems must be familiar to the user and acceptable
to him, and that they are an improvement only if they better serve
the intelligence operator and the producer of intelligence.
5. We consider as an important requirement the continuous
development of compatible data elements, computer programs
and processing equipment not only for large integrated systems
but also for the optimal operation of manual or specialized
automated systems. At the same time, standardization can become
a fetish which overlooks the unique requirements of individual
agencies and which limits creativity and the analytical power of
automated systems. When this occurs productivity is reduced
rather than increased. Our objective in standardization has been
productivity rather than unanimity.
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
S-E-C-R-E-T `"0
ANNEX F
6. Although the design and maintenance costs of automated
data handling are high, the processing costs are low. The effort
to control the level of expenditure of the Agency in the face of rising
cost of men and materials and the increasing use of expensive
technical methods of collection has given us a strong incentive to
adopt new techniques whenever they offered economies. In recent
years, four large automated systems, two in advanced stages and
two in the early operational stages, have been undertaken by the
Agency. Withal, the cost of information handling continues to rise.
Processing economies per unit have been more than offset by
increases in the volume of information we process. Our present
planning contemplates that total expenditures will level off after
1970, but this planning is based upon the evolution of present systems
rather than on a jump into large automated community systems. We
can spend more for automated systems than we have. But we can
do so only at the expense of severaly declining returns (cost-benefit
ratios).
II. Summary of Major Information Processing Projects
7. The intelligence information processing activities of the CIA.
have been undertaken to. support and extend analytical effort rather
than as ends in themselves. They have been limited by the conventions
of the intelligence environment (such as compartmentation and need-
to-know), but they have consistently sought to provide analysts with
as complete and relevant a body of information as was available within
these limitations.
8. Agency ADP activity can be divided into several major
categories; 1) general information storage and retrieval projects --
the library-like systems which are repositories of a spectrum of
information broadly relevant to the needs of analysts and operators,
2) special information storage and retrieval projects -- the personal,
organizational, or speciality files which serve a single analyst or
a group of analysts working in a common field, 3) data reduction
systems -- applications characterized by a large body of data which
must be perused in its entirety to select a small parcel of relevant
information, 4) data transformation systems -- a process to change
data from the form in which it is originally received to an acceptable
form for another analytical operation. Although a particular Agency
project may fall in more than one of the above categories this
categorization is useful for discussion of processing activities.
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
S-E-C-R-E-T lao~
General Information Storage and Retrieval Projects
9. The CIA has been engaged over the past 6 years in a
major effort to redesign its central information storage and retrieval
activities. This effort has focused on the application of modern
automated. techniques of information handling and on replacing a
multiplicity of central information systems, compartmented. by
collection source and. security classification, with a more generalized
and, flexible system. The design objective of this effort was to
simplify the accession and classification of intelligence information
and to improve significantly the value to the analyst of newly
structured. all-source files,
10. The design effort is essentially complete and. the Agency
has begun to implement several major elements of the new system,
including the development of all-source files, the organization of
central reference analysts by region, document indices produced
in machine readable form, and automated. computer search and
retrieval designed to operate in either a batch or on-line mode.
The major remaining element, a complete set of computer programs
(CAPRI) to support general information systems will be completed
this summer. Thus, we expect that it will be at least 18 months
before a preliminary operational evaluation of our general
information system can be achieved.
Special Information Storage and Retrieval Projects
11. Special purpose information storage and. retrieval is
a product of the increasing d.ivision of labor in intellectual and
analytical activity. It is also responsive to the explosion of know-
ledge and information which has made it increasingly difficult to
date for central systems to meet particular needs. The result has
been a growing requirement to classify and organize data to meet
these special needs. Specialization tends to limit the general utility
of the files, however, so that frequently the only consumers are
the individuals who process and generate the information. Even
where there is a wider potential audience, the data and. the research
methodologies used tend, to be meaningful only to the specialist and
to reside with him or under his control. Wherever possible, therefore,
the intelligence product d.erived from special systems is incorporated
S-E-C-R-E-T
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RD P82M00097R002600090004-1
der'
S-E-C-R-E-T
into general information systems; in any case, their existence is
made known to others who might find them useful. In the past, we
have sought to incorporate specialized projects into general systems.
Usually, however, the volume of data and. the complexity of their
manipulation have frustrated the attempts of general system
builders to record. more than the fact that special data systems
exist, their relevant product, and where or with whom the systems
resid.e.
12. Although, or even because, they are narrowly focused,
we believe that specialized information systems can be highly
productive, and justified both from the point of view of utility
and of cost. Appropriate steps have been taken within the community
to provide access to these systems in order to minimize duplication.
These efforts are stimulated by common professional interests
and. interaction and by the maintenance and. publication of the USIB
File and. Program Catalog. The latter provides a list of files and
computer programs for the special purpose systems which have
been automated..
Data Reduction Systems
13. Data reduction systems are characterized by the
manipulation of large bodies of collected data (in analog form, or
converted, to digital form), the analysis and extraction of significant
elements in the original data and the incorporation of these data
into appropriate analytical files. Data reduction systems are
usually coresident with special information storage and retrieval
systems.
14. Most data reduction in the intelligence community results
from the manipulation of data produced by technical collection
devices. The product is often not usable intelligence in and. of itself;
but raw material for further analysis. Frequently, both raw material
and. analyzed intermediate information are maintained. in machine
files, which are copied and used. by other intelligence agencies. The
data sets and computer programs are less well advertized in the
community than are those for special purpose information processing
programs, but they seem to be ,as widely known and exchanged
nevertheless.
S-E-C-R-E-T
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
S-E-C-R-E-T
Data Transformation Systems
15. Data transformation is difficult to define with precision,
because it may range from simple input functions such as key-punch,
to optical character reading (OCR) systems, to elaborate analog-
to-digital converters and automated printing systems. CIA activity
in these systems has ranged from modest in the OCR field to
substantial in analog-to-digital conversion, high precision plotting
and graphics, and automated printing. The Agency has participated
in government wide efforts to coordinate activity in the OCR and
printing fields. It has made known to others both within the outside
the government its procedures in precision plotting and automated
cartographic applications despite the sensitivity of certain applications
in this area. The extreme sensitivity and essentially developmental
nature of much of the Agency's activity in high-speed, analog-to-
digital conversion has barred widespread sharing of the results.
Nonetheless, the agencies with processing responsibilities in this
field have been kept informed.
Research and Development
16. Three years ago the CIA. drew together into its Office
of Research and Development a wide range of activities previously
pursued in separate components. This office was charged with
exploring information handling techniques at or near the edge of the
state-of-the-art, and with developing techniques within the state-
of-the-art which cannot be tested in a production environment. At
present, a modest effort is being made to identify equipment and
techniques which may augment intelligence information processing
in the future and to explore in a laboratory environment the
feasibility of incorporating them into the Agency and the community
as soon as they are cost-effective. This activity is particularly
important to the Agency because the intelligence production
components have had little opportunity for experimentation under
the twin pressures of rapidly rising flows of information from
new collection devices and growing demands to produce increasingly
accurate finished intelligence on an even wider range of vital issues
and areas.
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
-r;-U-K-ft-1 %we
Defense Intelligence Agency
SUMMARY
A. APPROACH TO THE PROBLEM OF INTELLIGENCE INFORMATION HANDLING
The DIA is making a concerted effort to approach improvement of
DoD intelligence from a comprehensive system point of view. This
involves addressing the information substance of user requirements,
defining the intelligence structure used to fulfill requirements as an
information system or network, identifying inadequacies in the system
in this context, and then proceeding with improvements or remedial
measures.
DIA's recent design of a long-range threat assessment program, for
example, represents a major effort at improvement of estimative tech-
niques and capabilities. This typifies the problem-oriented approach
to system improvement in which performance of a critical function is
enhanced first by improving functional system design, then by embracing
the techniques and technology which are optimum for the function. The
experimental Executive Management Planning and Control System (EMPAC)
views DoD intelligence as a system of interacting processes and opera-
tions on data streams and seeks to optimize the performance of the entire
system by addressing its parts in the context of the whole. DIA partici-
pation in analyses initiated by the Bureau of the Budget, Director of
Central Intelligence, and Secretary of Defense further contributes to
this objective. The Director, DIA is actively considering organizational
adjustments which will serve to focus and intensify the Agency's capa-
bilities and efforts in this approach to intelligence and information
handling problems.
B. DoD INTELLIGENCE DATA HANDLING SYSTEM
The worldwide DoD Intelligence Data Handling System (IDHS) consists
of the facilities, equipment, special data communications, procedures,
and personnel which provide technical and operational intelligence data
handling capabilities in support of general intelligence production in
U.S. Military Commands and organizations. It currently includes 49
computers installed in 21 locations.
This system has been developed on an evolutionary basis. The guiding
principle has been that the system must be expended in small enough seg-
ments so that they can be implemented before'"the external environment makes
them obsolete, and so that their impact and effectiveness can be assessed
before additional developments are implemented. To-that end, the system
is under frequent evaluation, both from a technical and a user point of
view.
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
S-E-C-R-E-T Nfto
Since its inception in 1963, the objective of the worldwide DoD IDH
System has been to constantly improve intelligence data handling through
the development of a system of mutually supporting facilities. Important
steps which have been taken toward this end include:
(1) The development of families of computer programming systems
(e.g., the Formatted File System).
(2) The standardization of data elements and codes.
(3) The widespread exchange of data bases.
(4) The availability of technical assistance to commands in
their early stages of automation.
(5) The generation of management plans which delineate the
respective responsibilities of DIA, the Military Depart-
ments, the Unified and Specified Commands, and various
user commands.
periodic evaluations of the results of the IDHS efforts to date dis-
close that ADP is making its most significant contribution to military
intelligence in the following broad functional areas:
(1) Exploitation of photography.
(2) Exploitation of ELINT.
(3) Targeting.
(4) Missile Trajectory Computations.
(5) Foreign Ship Activity.
(6) Various activity files pertaining to Viet Nam.
(7) Orders of Battle.
(8) Dissemination of Intelligence Reports.
(9) Storage and retrieval of intelligence reports.
Looking toward the future, major improvement efforts are underway in
a number of areas. Basically, these efforts center on the development of:
(1) remote access, time-sharing systems to enhance support to intelli-
gence analysts; (2) a network of secure digital data links to permit rapid
exchange of information among the various commands; (3) automatic input
S-E-C-R-E-T
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
or, S-E-C-R-E-T
devices and techniques such as the optical character reader; and (4)
advanced training programs in information science for personnel at all
levels. Some of the major projects currently underway are listed below
and described more fully in ATTACHMENT B:
Project ANSRS (Analyst Support and Research System).
Project COINS (Community On-Line Intelligence System).
Project COLEX (CIRC On-Line Experiment).
Project VASS (Visual Analysis Subsystem).
Project PACER (Program-Assisted Console Evaluation).
Navy Ocean Surveillance System.
In general, up to the present the DoD Intelligence Data Handling
System has made its most significant contribution to intelligence in the
areas of basic intelligence production (installations files, orders of
battle, etc.), support to the exploitation of photography and ELINT,
certain types of current activity analysis (foreign ship activity), and
the storage and retrieval of reports and documents. It is beginning to
make a significant contribution in the areas of direct support to intel-
ligence analysts, and management of intelligence operations. To date,
it has not made a significant contribution in the areas of estimates and
warning. It is these areas to which increased attention must be focused
in the immediate future.
Funding and personnel data relative to the operation and maintenance
of the DoD Intelligence Data Handling System for FY 1969 through FY 1973
are included as an attachment to this summary.
C. THE INFORMATION SCIENCE CENTER
The Information Science Center is being established to develop and
conduct specialized courses of instruction in the application of informa-
tion science to specific intelligence problems. Initial courses (sched-
uled for implementation in January 1969) will deal with the intelligence
planning, estimates, and warning areas. Subsequent courses will address
other specific categories of intelligence problems.
D. INTELLIGENCE EXPERIMENTATION CENTER
The Intelligence Experimentation Center is a developmental effort
aimed at accelerating the application of modern methods, techniques, and
equipments to.the improvement of intelligence processes. The Center will
S-E-C-R-E-T
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
S-E-C-R-E-T .r
use an intensive and direct experimental approach. It will design and
conduct both laboratory and operational tests of innovative methods and
technologies, and it will prepare implementation plans for the systematic
introduction of demonstrated improvements into the operating intelligence
system in DIA and in the intelligence elements of the military commands.
The Intelligence Experimentation Center will be activated in FY 1969 and
will respond to guidance and recommendations of the Director of Defense
Research and Engineering.
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
TOA ($ in millions)
FY 69
FY 70
FY 69 thru 73
DIA
$14.1
$15.6
$ 72.2
ARMY
8.0
7.6
39.7
NAVY
13.7
11.7
57.3
AIR FORCE
28.0
24.8
116,6
Total TOA
$63.8
$59.7
$285.8
SOURCE: FYDP, January 1968
WORLDWIDE DoD IDHS - MANPOWER
FY 69
FY 70
FY 71
FY 72
FY 73
DIA
ARMY
NAVY
AIR FORCE
1,273
1,268
1,268
1,268
1,268
Total
2,667
2,674
2,674
2,677
2,677
SOURCE: FYDP, January 1968,
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
a-.C.-I..-K-J' -1
irk
SECRET 9 April 1968
Summary of Major Intelligence-Related
Information Handling Activities
National Security Agency
INTRODUCTION
More than 85 computers are used in the NSA computer
complex to organize the SIGINT data received daily in
various forms from a world-wide collection effort. NSA
information handling processes and systems are fused
throughout the entire SIGINT production cycle, from the
point of intercept through delivery to the customer.
Accordingly, many NSA information handling activities
are inter-twined with the SIGINT production processes
and frequently cannot be separated. The production
processes are oriented towards target areas or technical
fields, such as ELINT.
Many of the systems used in the SIGINT process.are
software computer programs which use one of a number of
"standard" computer configurations. These "standard"
configurations include the IBM System 360/65 or the
CDC 6600, or the agency developed computer complex HARVEST.
The prolific use of computers in the SIGINT process
is complemented by an in-house endeavor to write the
programs in an "open shop" environment. The majority
of the SIGINT software development is carried out in this
environment. In fact, the most useful software develop-
ments originated with the problem solving people in the
production offices. Programming personnel in the computer
complex have devoted their time more profitably to the
specialized and more difficult areas of computer system
software. "Open shop" programming has advanced the
computer art at NSA because the user is confronted with
25X1 25X1
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
the hard realities of his problem as it relates to the
computer and, is apt to resolve his own problem in his
own way.
The NSA Summary of its information handling activities
is separated to show some general support systems
(RYE/TIPS and VIDEOFILE), natural language processing,
off-line storage and retrieval systems, data preparations
and conversion, communications, and research and development
applications. Not included in this summary are the manual
information handling systems in use in NSA.
RYE/TIPS/COINS
The most significant NSA systems related to community
information handling are the RYE-TIPS/COINS Systems. The
RYE system is a large scale, remote access, general purpose
digital computing system. It is a real-time system which
in this context is defined as the processing of transactions
as they occur or as the processing is desired. RYE is
designed to handle a class of problems which cannot be
satisfactorily handled by any other computer system. These
problems are characterized by small size, a need for immediate
reaction, or a need for quick file reference. The system is
designed so that an optimum number of processing tasks can
be accomplished in minimum time consistent with the urgency
of requests. Such problems can be handled more economically
and efficiently by a centrally located large scale computer
to which many users have access through remote terminal
stations rather than by traditional batch processing methods
or a scattering of small scale computers dedicated to
specific uses or users.
The system is widely used throughout the NSA Production
Organization by cryptanalysts to process encrypted traffic
with diagnostic and statistical programs. In addition to
many specially written programs for specific projects, a
2
S-E-C-R-E-T
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
5-r:-C:-K-E-'1'
large library of general utility programs are available.
The RYE System has shown its operational effectiveness
by the phenomenal growth in its utilization. Individual
processing tasks increased from approximately 32,000 per
month in January 1966 to approximately 80,000 per month
in December 1967.
The Technical Information Processing System (TIPS)
implements the concept of a remote access query, retrieval
and data management capability. It operates on the RYE
System. In TIPS, files of diverse and, dynamic information
pertinent to SIGINT collection, processing and analysis are
maintained by appropriate offices. The files are accessible
for interrogation by all approved users from any one of
about 100 remote terminals.
On 1 June, the TIPS System will be connected to the
Remote Access Computer System of the Air Force Security
Service, Kelly AFB, Texas, for interrogation of files
on each system by analysts at each location.
Based on our experiences with RYE/TIPS operations, NSA
has been actively engaged in developing the COINS concept
in conjunction with others in the intelligence community.
Currently, on-line tests are being conducted between NSA
and DIA.
The utility of a VIDEOFILE technique is currently under
NSA investigation as one approach to its document information
storage and retrieval problem. This technique permits the
handling of document images in analog form on magnetic tape.
They are stored as TV images and are displayed on a cathode
ray tube console as a TV picture. The system can retrieve
images,of the original documents and provide either a visual
display or hard copy. Seven reels of magnetic tape can store
the contents of 150 four-drawer file cabinets.
3
S-E-C-R-E-T
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
S-E-C-R-E-T =.rr
NSA has embarked on the first phase of a three phase
program to install and operate a VIDEOFILE storage system.
The first phase is a pilot experimental system which will
be ready for demonstration by the latter half of 1969.
Follow on phases are planned but are contingent on the
success and. the utility of the initial phase.
S-E-C-R-E-T
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
.7- CI- V-tc- C - 1
j..r
SECRET
A second and the largest class of systems are those
which store partially structured information in formatted
files. Elements of information are recorded in natural
language but tagged so that the machine may recognize the
nature of the element.
A third class of systems is designed to collect
information on rigidly structured organizations (e.g.,
military or governmental) by placing new pieces of
information into a computer memory in hierarchical
locations which parallels the hierarchy of the organizations
themselves. By reflecting the structure of the organizations
in the machine file structure, information retrieval and
report publication are greatly simplified.
S-E-C-R-E-T
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
The use of optical readers is one of several ways
NSA is attempting to improve its data preparation and.
conversion processes. NSA is using keyboard operated
visual display devices for direct machine record editing.
These devices make corrections directly into computer storage
media which eliminates extensive repunching of machine
record corrections.
In order to cope with growing volumes of communications
data, both for computer processing and message distribution,
NSA is actively planning an on-line computer-based data
handling system. Definition is underway of a computer-based
data handling complex capable of accomodating the current
workload, with sufficient growth capacity to facilitate
future expansions of the communication satellite data links
across the Atlantic and Pacific Oceans. An internal
distribution system is being installed for automatic
distribution of teletype traffic. This system will use two
computers at handling speeds of 400 characters per second..
S-E-C-R-E-T
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
ANNEX H
Some of the new concepts being explored in information
handling include many aspects of software, hardware, and
information system management. Two separate, but related,
mass storage implementations are planned. The first is
the development of a mass-storage capability (1012 bits)
to replace the NSA computer tape library for on-line use.
A mass storage system and host computer are being
implemented for this replacement. The existing library
is now in excess of 100,000 tapes with additional input
being received at an increasing rate. Retrieval, storage
and manual management of a library of this size is time-
consuming and expensive in manpower. Complete, on-line,
automatic management will be provided by a computer which
controls the mass storage devices. Concurrently, the
future large computer complexes planned will require mass
storage devices of their own. The second type of
implementation will provide several on-line libraries
to hold a large volume of file data, in a state of ready
access, to be used and generated by a single type of
computer system. Ideally, these data should remain within
the area of the user computer, thereby, removing the
burden of handling, storage, and documentation from the
tape library.
Further work is being devoted to make the computer
available to the analysts through a gr
rather than a teletype or typewriter.
Because of the steady "information explosion" in
page-printing output, an automatic computer photo-typesetting
system is planned to close the time-gap, . between completion
of data processing operations and the production of a
7
S-E-C-R-E-T
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
printed document as output. Higher speed processing and
volume reduction are needed. Typesetting devices are
now available that will produce, on film or paper, text
matter of high quality from the output of a computer.
Such a capability is being provided for the RYE System.
Other R&D tasks include: (1) development of a large
multiple input switch for computers for paralleled input
of data; (2) a combined mass storage and retrieval system
for digital, analog, and photographic forms of information;
(3) methods to relieve the demands on the processor and
input-output equipment for sorting and table look-up
operations by the use of associative or large memory
techniques; and, (4) search for solutions to the problems
of interconnecting many (i.e., ten or more) computer
equipments in a centralized automated computer complex in
order to provide a pool of computer capability.
8
S-E-C-R-E-T
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
 25X1 Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
Next Next 1 Page(s) In Doc
Sanitized Copy Approved for Release 2009/11/13: CIA-RDP82M00097R002600090004-1
ument Denied