COMMUNITY ON-LINE INTELLIGENCE SYSTEM (COINS), IMPLEMENTATION PLAN

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP80B01139A000100140007-1
• Release Decision: RIPPUB
• Original Classification: S
• Document Page Count: 41
• Document Creation Date: December 20, 2016
• Document Release Date: October 19, 2007
• Sequence Number: 7
• Publication Date: May 25, 1966
• Content Type: MF

Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
UNITED STATES INTELLIGENCE BOARD
NSA review
completed
CONNITTEE ON DOCUMENTATION
25 May 1966
MEMORANDUM FOR: Chairman, Coittee on Documentation
SUBJECT: Community On-Line INtelligence System (COINS),
Implementation.:Plan
1. Transmitted herewith is the Implementation Plan for COINS.
Drafts of this plan have been extensively reviewed by your Committee as
well as by each of the participating Agencies. It is assumed that this
plan will be approved since no objections were voiced during the review
process and recommended changes have been incorporated in the final
version.
2. Therefore, while awaiting final approval of this paper, the
COINS Committee will continue the necessary detail planning actions in
the following areas: (a) communication engineering specifications;
(b) message format and exchange procedures; (c) file specifications
and standards; (d) simulation and evaluation procedures. Additional
reports will be submitted as planning is completed in each of these
areas. This approach is being taken to insure meeting the implementation
date of 1 December 1966. However, actual implementation steps such as
file preparation, programming, procurement and installation of equipment
must await final approval of this plan and will require a five to six
month lead time.
3. Implementation of COINS I is not contingent upon the immediate
or simultaneous participation of all Agencies. Therefore, the fact
that at this time CIA has not selected its files nor defined its user
language should not delay approval of this plan for immediate
implementation. As soon as this information is provided it will be
integrated into the overall. system.
DIA review(s)
completed.
SECRET
Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
K.J L/ %.A U. L " I.
N TABLE OF COXNTS
CHAPTER
PARAGRAPH .
I TOPIC
PAGE
'
GENERAL
1
Purpose
1
2
Authority and Direction
1
3
COINS C _.romittee
2
4
What is TIPS at NSA?
3
5
Concept
5
6
Ultimate Objectives
5
7
Approach
6
8
Security
7
9
Restrictions
7
10
Recommendations
FILES
7
1
General
9
2
Files for COINS I
9
3
Categories of Files
10
4
Data Element Standardization
10
5
File Organization and Formatting
11'.
6
File Maintenance
U
7
Problem of Duplicate Information
in COINS
INTERROGATION PROCEDURE
11
1
Objective
14
2
Response Time Required.
11E
3
Expected Volume
124
1.
Interrogation Priorities
.
14
5
Scope of Interrogations
16
6
Interrogation Strategy
17
7
Assumptions
17
8
Type of Interrogations
18
9
Specific (?canned") Interrogations
18
.10
Users Language
COMMUNICATIONS
20
1
Commmanications Switch
23
2
Points of Agreement on Co]mnunlcation
23
a.
Transmission Rate
23
b.
Security
214.
Co
Standard. Comunications Code
(ASCII)
24
do
Length-of-Messages
24
as
Message. Transmission
24
f.
Time of Operations
24+
g.
Error Detection and Correction
(EDC)
24
i
SECRET
Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
CHAPTER
`'
PARAGRAPH
V"O
TABLE OF CONTENTS
(Cont'd)
TOPIC
PA
GE
COMMUNICATIONS (Cont'd)
h.
Transmission of Files
24
is
Communications & Cryptographic
Equipment
25
Type-of-Subscribers
25
3
Messages
25
a.
Type-of-Messages
25
b.
Message Formats
26
Method of Operations for COINS I
26
IMPLEMUTATION SCHEDULE, COSTS AND RESPONSIBILITIES
1
Implementation
29
a.
Target Date
29
b.
Constraints
29
c.
Costs
29
2
Responsibilities
29
as
General
29
b.
Communications and Cryptographic
Equipment
31
C*
Software
31
d.
Computers
32
3
Responsibilities of COINS
Committee
33
4
Evaluation
34
SECRET,
Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
TABLE NO.
I
Interrogation Priority Conversion Table: STATE/DIA
to NSA/NPIC
15
II
Interrogation Priority Conversion Table:
NSA/NPIC to DIA
15,
III
Advantages and Disadvantages of Different Interrogation
Techniques
19
IV
Estimate of Implementation Costs for COINS I
APPENDICES
30
I
COINS I
II
COINS II
35
37
III
Displays of the Interrogation Cycle
39
iii
SV CP TI~.~T"
Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
REFERENCES
1. DC ID Communications telligence Regulation, Effective 29 December
1959
2. Staff for the Community Information Processing Study (SLIPS)
Stage I Report I Volume I dated October 1963
NSA/C2 "NSA TIPS Reference Manual" dated May 1965
4. Letter from the White House, dated 15 June 1965, Memorandum for the
President, Subject: "U.S. Intelligence Conmiunity Capabilities for
the Handling of Intelligence Information"
5. Letter from the White House, dated 15 July 1965, Memorandum for the
Director of?Central Intelligence, Subject: "U.S. Intelligence
Community Capabilities for, the Handling of Intelligence Information
USIB D-39 7/11 dated 24+ September 1965, Subject "U.S. Intelligence
Community Capabilities for the Handling of Intelligence
Information"
USIB M-5 dated 7 October 1965, USIB Approval of CODIB Report Contained
in Reference 6
8. DIA/IDHS FFS Retrieval and Output Techniques Volume I dated
15 October 1965
9. DIA/IDHS FSS Retrieval and Output Techniques, Volume II dated
15 October 1965
l0. "RYE.l+90 Ref r c " t 1.965, Published by NSA Office of
Training
11. USIB D..39 7/13, dated 5 January 1966, Subject "C amnittee on
Documentation Report of Task Team IV (Installations).
12. CODIB Task Team V - Biographies, Subject FINAL REPORT, dated 1 February
1966
13. JCS 222/935. Paper dated 14+ March 1966, subject: "Expansion of
Washington Area High Speed Facsimile. System"
14. DIA Draft Working Paper on ISIC Query Language
No date
SECRET
Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
I_J.U u `L' I.
V
IMPLEMENTATION PLAN FOR COINS I
1. PURPOSE
The purpose of this paper is to provide the management of each of the
participating Agencies with the detailed information necessary to make a
judgement with respect to the implementation of COINS I. A detailed
examination of this paper will indicate that the: (a) linking together of
sel
t
d
t
ec
e
compu
er systems is technically feasible by 1 December 1966, and
Chiefs of the participating Agencies will constitute a commitment on the
part of those Agencies to provide the resources necessary to implement COINS I
by the aforementioned date subject to the constraints listed in Section V.
para lb.
2. AUTHORITY AND DIRECTION
This plan. was prepared in response to the White House Memorandum, dated
15 July 1965 (See Reference 5). Specifically, this letter directed that
Recommendation No. 2 in Reference 4 be implemented. Following is extract
of this recommendation:
"Recommendation No. 2: That the Technical Information
Processing System (TIPS) project, now underway within the
National Security Agency, be expanded to include
participation by other member, Agencies of the Intelligence
Community in an experimental, operating system constituting
a first step toward inter-agency (and inter-building)
information handling. Since results should be sought from
the experiment as promptly as feasible, the participation
of other Agencies should be achieved by September of 1965;
thecapability for extensive handling
problem should be available 'in the community-wide
system by the summer of 1966; and by the summer of 1967
if should be possible to exchange outputs from various
mechanized sources in the fashion pioneered by the TIPS
project.
(Only through such experimental operational trials can
the Intelligence Community come to grips with the wide variety
of program problems involved, including those of security
compartmentation, the encryption of communications between
the computer/information base and the user locations, and
SECRET
Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
other blems.) In order to make suchtrial effective,
it may be necessary to expand the scope of the information
maintained in the TIPS system and, if so, this should be done
with caution as to the total amount of material thus added.
The intention should be to establish a system that will, in
fact, be used by workers in at least a few Agencies as a
better way to meet day-to-day tasks; however, the system
should be regarded as experimental and there should be no
attempt to insure that in its experimental form its operation
can be economically justified."
3. COINS O NMITTlEE
a. The Community On-Line INtelligence ystem (COINS). Committee was
established in September 1965,7under the auspices of the-Committee on
Documentation, United States Intelligence Board (COD1 ). The Committee
was charged with the responsibility of developing the plan to implement
Recommendation No. 2 .in Reference 4. The membership of this Committee is
as follows:
Chairman
Secretary
b. Acknowledgement
Following by organization is a list of the many people to whom the
members of the COINS Committee are indebted for considerable technical
assistance and advice they have provided during the preparation of this
paper. The members of the COINS Committee however.accept sole responsi-
bility for the contents of this paper.
SECRET
Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
4. WHAT IS TIPS AT NSA?
a. NSA management became increasingly aware of the mounting volumes
of technical information (i.e., information with respect to SIGINT
targets and resources) within the SIGINT Community as well as the
inadequacies of existing methods for storage, correlation and retrieval
of this information. In 1963, it was decided that there was an urgent,
need for a radical approach in this area, taking maximum advantage of
existing automatic. data processing technology. As a result, NSA has
been engaged, for the past three years in planning, developing and
implementing of the Technical Information Processing System (TIPS) in
a concerted effort to improve the assembling.-processing and retrieving
of formatted technical information. This project is being accomplished
in three phases, namely, TIPS "PILOT", TIPS I and TIPS II. The first
two phases are being accomplished in-house; the latter phase, however,
is under study by various components of the NSA/R&D organizations with
some contractual assistance.
(1) TIPS "PILOT" - This is an experimental retrieval system now
entering its final stages of. implementation using the available UNIVAC
490 remote access system (RYE). The existence and availability of the
RYE system is one of the key contributing factors which has permitted
the successful implementation of TIPS "PILOT". This system calls for
the centralization of a selected set of thirteen formatted files, con-
taining approximately 38 million characters which are representative of
the types of files to be accomodated in future generations of TIPS.
Information contained in these files is of interest only to members of
the cryptologic community. The pilot system is designed to meet a
limited number of operational requirements, and therefore it is available
24-hours per day, seven days aweek. The main objective is to enable
NSA technicians to learn and gain operational experience in designing,
developing, organizing, programming and utilizing an on-line near real
time information retrieval system. Some of the more provocative questions
which need to be answered are: Is such a system required and is it worth
the cost? Will the NSA organizations u
updated more. fre uent UNCODED
detailed description of
an can be made available by NSA to any authorized
organization providing they have the proper security clearances (i.e.,
SECRET SI).
(2) TIPS I (July 66 - Dec 67)
(a) Plans and studies are now underway for the initial
expansion and. refinement of TIPS "PILOT" to be known as TIPS I. It is
evident that the existing RYE complex cannot handle the anticipated
work load resulting from TIPS I. Therefore, implementation of TIPS I
SFCRr,T
Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
kJLJXLIiVL 1.
will not begin until after the upgrading of the current RYE system has
been conroleted, which is tentatively scheduled for the second quarter
of F^.t67 (i.e., Oct - Nov 66). This interim' period will be used to
accomplish the necessary detailed system planning for implementation
of TIPS on the RYE system. it is estimated that the implementation of
TIPS I will require approximately 150 man years of effort spread out
over 12 - 18 months.
(b) A generalized file maintenance program and a users
language with a report format generator will be included in the system's
software for TIPS I. Inclusion of this software will make the system
more flexible and responsive to the constantly shifting information
display requirements of the users. This action will reduce but not
eliminate the need for open-shop programmers in TIPS I.
(c) The following tentative estimates are provided for
(1) Files currently contained in TIPS "PILOT"
combined with those selected for TIPS I will contain approximately 160.-
200 million characters and be divided into roughly 60 - 70 different
files.
(2) There will be approximately 15,000 to 20,000
inputs per day with the majority of them occurring during the .,normal
duty hours, Monday thru Friday.
(a) The system will be designed to (a) provide an
average response time of 15 minutes for short term interrogations and
(b) handle all procedural requests within the system on a 24-hour turn
around basis.
() There may be approximately 150 - 175 remote
stations involved, some of which may be located at the Service,Headquarters
(i.e., AASA, AFSS and. NSG), as well as other Z.I. locations. Soave of these
remote stations may be equipped with visual display devices.
(3) TIPS ~T~TO (r - i A This is a third
generation of TIPS and is currently under study by various components of
the NSA R/D organizations with contractual assistance. The data base
for this system will. contain approximately 500 million characters with
required. response time of less than five minutes for short term inter-
rogations.,- This system will require the development,of basically new
approaches in file organization and search strategies coupled with
technclogical advances in computer hardware. In addition, it is envisioned
that major overseas sites will be capable of directly interrogating and
receiving answers from the system independent of human intervention.
b. The long range objectives of TIPS are to:.
(1) Establish a complete. mechanized data base with respect to
SIGINT targets and resources which is routinely updated in a tidy
fashion and which provides NSA technicians and managers at all organi-_
zational levels with current and accurate information.
SECRET
Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
~/ it t/- ~. iJ Z
(2) Provi%,the basis for the eventual imp entation of the
SIGINT
Carmnand and Control Complex (SCCC) in FY70 - F771. The implementation of
this concept is dependent upon the successful establishment and maintenance
of the mechanized data base described above . The SCCC must have, in addition,
two characteristics which TIPS will not provide. Using the TIPS data base,
it must:
(a) Automatically assess incoming information and alert the
appropriate organization of an abnormal situation or, in routine situations,
automatically make decisions leading to appropriate actions;
(b) Provide answers to interrogations which indicate possible
alternatives for action, and give them some weighting according to general
criteria so that optimum actions are easy to select.
5. CONCEPT
a. In accordance with the White House letters, subject "U.S. Intelligence
Community Capability for the Handling of Intelligence Information" dated 15
June and 15 July respectively, a secure network of remote information re-
trieval computer systems is being planned within the USIB Community, i.e.,
NSA, DIA, STATE and CIA (i.e., Information Processing Division of the
Interpretation Center - EPIC) and the CIA Computer Center. The title being
used for,this project is Community On-line INtelligence Crstem (COINS).
b. Using COINS, each participating Agency will be able to remotely
interrogate its own file(s) as well as selected files of the other partici-
pating Agencies. This concept calls for each Agency to maintain its files in
its own remote information retrieval computer system and for the computer
system of each Agency to communicate with one another via secure data links.
This approach eliminates the necessity for one Agency to serve as the central
repository for the, entire Intelligence Community. Instead., each Agency will
concentrate on establishing and maintaining those files for which it is
responsible or best suited.
c. This system is not designed to solve all of the inter-agency problems
in the Intelligence Community with respect to information retrieval or infor-
mation exchange. Specifically, COINS is aimed at formatted files, as opposed
to scanning and. selecting desired portions of narrative text files. Formatted
indices to narrative text files, biographic dossiers, finished intelligence
files, micro-film or video-film images or documents may be retrieved through
COINS. For example, if an interrogation of COINS reveals that the desired
biographic dossier or peice of finished intelligence exists in the files of
another Agency, the secure high speed facsimile system (network) planned for the
Washington D. C. area could be utilized for the inter-agency transmission of
the desired report or dossier. (See Reference 13). Although some files may
contain narrative text the desired information will be extracted based on a
formatted . or key field.
6. ULTIMATE OBJECTIVES
a. Reduce the duplication of effort by eliminating the necessity for
maintaining and supporting a multiplicity of EDP programs and formatted
files of similar content by direct inter-agency computer communication.
SECRET
Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
 Approved I-or Kelease 2001/10/1y : (IA-KUF- ObO11:3 AUUU1UU14UUU/-1
".LJ t.111.JJ J.
b. Improve the com City's capability to exploit t1 ever-increasing
volume of intelligence by improving timeliness in the processing, maintenance
and, distribution of finished, semi-finished.,,and key intelligence information.
c. Provide a high degree of flexibility in managing, selecting,
collating and distributing intelligence information.
d. Improve the opportunity for the effective utilization of finished,
semi-finished, and key intelligence information by making it readily
accessible to technicians at various consumer and intelligence producing
Agencies in a useful -time frame.
e. Establish a basis for designing and constructing a more sophisti-
cated, dyndmic intelligence network in the future.
f. Provide for more effective and efficient utilization of equipment,
manpower and timeo
g. Develop the security requirements and controls necessary for
dynamic intelligence information exchange.
7. APPROACH
This concept will be implemented in two phases, namely:
a. COINS I (Dec 66 - Jun 67)
(l.) In general, this plan calls for the use of existing computer
systems or an upgraded version of existing systems at NSA, CIA, NPIC and
DIA. The State Department will participate as a consumer in this phase
by having a remote station tied into the DIA computer system. The
primary objective of the pilot phase will be to gain some experience with
respect to the technical problems associated with establishing, using and
maintaining a secure network of remote access computer systems within the
intelligence community. A parallel objective is to improve the,inter-
change of information within the'Intelligence Community.
(2) Although unsophisticated and. primarily designed. to determine
feasibility, COINS I is an important and necessary first step toward the
development of a larger and more complex community-wide system in the
future. Implementation of COINS I is a challenge to the U.S. Intelligence
Community for it will necessitate very close cooperation among all parti-
cipating Agencies. If these Agencies cannot work together to implement
a simple, straightforward system such'' as COINS I, then it certainly will
not be possible to implement more sophisticated community-wide systems
in the future.
b. COINS II (n..68)
Plans with respect to this phase will be influenced by each
Agency's internal experience (a) in implementing COINS I, and (b)
developing their own on-line analyst support systems. As a basic minimwn
COINS? _will require a self-s. mkt access computer system
SECRET
Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
 Z Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
_ 0.G%AIIEIv
at each Agen.ry l through a CIA installed and staffed cmmunic jQns
sore and forwardinked switch. The network xil 'eve op common interrogation
an c o tSe
8. SECURITY
a. Prime Factor
Security is a prime factor of consideration in establishing
(1) Initially, in COINS I all computer complexes, remote stations
and communications will be protected to TOP SECRET SI. This approach will
avoid the multi-level classification problem. Any installation may
operate a computer complex with multi-level classification or multi-level
outstations providing such use has been approved in accordance with
Reference 1.
(2) However, COINS II must be capable of handling the multi-level
classification problem. In this phase, a computer and its remote outstations
will have to be located in a secure ar,a consistent with the highest classi-
fication of information it is authorized. to receive from the system.
b. Responsibilities in COINS I
Each Agency is responsible for its own internal security.
Specifically, it must insure that software and hardware built into its
computer system afford the degree of security required. For example, each
Agency is responsible for insuring that:
(1) Files are not accessed by unauthorized external organizations,
either for the purpose of interrogation or file maintenance.
(2) Responses are delivered to the appropriate remote station
consistent with security requirements.
(3) The classification of all information introduced into the
inter-Agency computer net is no higher than TOP SECRET SI.
9. RESTRICTION
Participation in COINS I will be restricted to the Agencies indicated
insure adherence to the implementation schedule. The participation of
additional organizations during this period would only present more problems
and. perhaps delay implementation.
1C. RECOMMENDATIONS
a. A series of specialized working groups should be established under
CODIB to develop data standards for the United States Intelligence Community.
These works groups should cover such areas as (1) rsonalities -
biographies (2order-of-battle (3) installations (4rgeographic, etc.
Wherever possible the charters of existing CODIB Working Groups should be
expanded to include this function.. (Reference Chapter III para 3 and
Chapter V para 3).
7
QV CD i 'r
Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
b. Immediate ste"PLf should be taken to officiallytify appropriate
organizations (i.e., DOD, STATE, CIA, NSA, etc.) that CODIB will be the
organization for:
(1) Developing data standards for intelligence activities for
Agencies which are members of USIB.
(2) Coordinating data standardization efforts within the United
States Intelligence Community with the Bureau of the Budget.
c. Agencies participating in COINS I develop the specification
for a common, computer independent users language for use throughout
the Intelligence Community in COINS II for remote interrogation and file
maintenance. These specifications to be based' on the operational experience
gained in COINS I.
SECRET
Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
OJP-d A IXJP-4 %M%#
II
FILES
1. GENERAL
According to Recommendation No. 2 of Reference 1+, initial efforts
of the Intelliaence Co unity were to be directed towards the
problem. Therefore. 7dicated in paragraph 2
e ow, a number of machinable, files have been
included in COINS I. However, two impor an factors have limited this
approach:
a. As indicated in Reference 6, the majority of the biographic files
are not in a machineable format. This fact is further substantiated in
CODIB TASK TEAM V Final Report on Biographies. (See Reference 12).
b. The kinds of files in any such system must be diversified since
not all organizations are interested in the same kinds of information.
.For example, the interest of State and NPIC differ widely.
2. FILES FOR COINS I
The number and types of files to be included initially in COINS I is
not particularly important, since the principal objective at the outset
is to implement the system and. work out the technical problems. Once the
system is in operation, the number and variety of files can be expected
to increase rapidly. The files which each Agency has decided to make
available to the Intelligence Community through COINS I are listed below.
Detailed, descriptions of these files have not been included in order to
hold down the classification level of this report.
(1) NPIC - Target Brief File (This is a restricted sub-set of
total file)
(2) CIA Computer Center - Files to be nominated at a later date.
25X1
25X1
9
Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
 C T1 4-N T T r-. i
Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
Air Order-of-Battle
Missile Order-of-Battle
Radar Order-of-Battle
Air Defense Order-of-Battle
Military Personalities
Viet Nam Activity File (Only the most recent 30 days)
3. CATEGORIES OF FILES
a. There are two general categories of files which may be found in
each Agency's system:
(1) Those files which are available for use by other members of
the Intelligence Community.
(2) Those files which are not available for use by other members
of the Intelligence Community. Each Agency is responsible for building
the necessary safeguards into their own computer system to prevent
unauthorized access to these files by other Agencies. For example, in
TIPS at NSA there are a large number of files which are reserved only
for internal use of the cryptologic community. These files will not be
made available to COINS.
b. DIA has indicated that all of the files in the ISIC/IBM 1410
on-line system will be available for interrogation by other COINS members.
4. DATA ELEMENT STANDARDIZATION
a. As soon as this plan is approved appropriate steps will be taken
by the Committee to establish standards for the files in COINS insofar
as it is practicable. However, it must be recognized that it may not
always be feasible to achieve ccmplete standardization for all data elements
for all Agencies in the Intelligence Community. For example, an Agency
may have already adopted certain standards for its internal operations
and to alter them would require extensive changes in: (1) file.formats
and contents, (2) computer programs, (3) input/output formats, (4) directives
and procedures, and (5) training courses.
b. The following rules have been established with respect to data
element standardization in COINS.
(1) Whenever standardization cannot be achieved, translation
tables will be constructed to translate between standard and non-standard
data .lements. In this event the rreequestin Agency is responsible for
(a) translating outgoing data elements of interrogations into the terms
of the receiving Agency and (b) for translating incoming data elements in
the response into their terms.
SECRET
Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80B0l 139A000100140007-1
(2) Standards adopted for the files it COINS will not be binding
on the other machine files of each Agency which are not part of COINS.
However, these data standards will be forwarded to CODIB for final
coordination throughout the Intelligence Community.
(3) Each Agency will be responsible for correcting and maintaining
the information in their files in accordance with mutually agreed upon
standards.
(4) Standard data elements approved by USIB and adopted by the
Intelligence Community will be used wherever such exists. (See Reference
il).
c. There is an urgent need for the standardization of data elements
within the U. S. Intelligence Community and such standardization efforts
should be accomplished. by a working group under USIB (or under CODIB)
with representation from each member Agency. This working group in turn
should be responsible for coordinating the Intelligence Community requirement
with the Bureau of the Budget which is charged with developing these standards
for the entire U. S. Government.
d. Geographic information in some form or another appears to be a
common thread running through all files. As a result, some geographic
standards are also needed in the U. S. Intelligence Community, e. g.,
coordinate representation, urban area code, geographic sub-division, etc.
5.
FILE ORGANIZATION AND FORMATTING
Each Agency is responsible for:
a. Organizing and formatting its own files.
b. Keeping other Agencies informed of any significant changes in file
format, accuracy or content. In this instance, content does not refer to
all .routine file maintenance changes, but rather
cats :ori s UNCODED
c. Converting the data elements in its file to the accepted standards.
6. FILE MAINTENANCE
The system will not permit an Agency to update the information in the
files of another Agency (i.e., NSA cannot update or modify information
contained in the DIA files, and vice versa) unless duly authorized on each
file affected.
7. PROBLEM OF DUPLICATE INFORMATION IN COINS
a. Goal
One of the major goals of this system is to reduce the necessity
~-r11 --
.~V.CR 1i , 1
Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
 !~1 T. !V T 7~ r-rti
Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
of maintaining duple banks of information. It isoubtful that
complete elimination of duplicate information banks can be achieved at
this time. Duplication may be required for economic reasons.. The
trade-off between maintenance of duplicate files and th
v
e ser
icing of
requests by the individual Agency against its copy of a file as opposed
to the establishment of a single copy of a file and the servicing by
one Agency of all requests against that file is not obvious at this time.
The relative efficiency of centralization depends on the relationship
between maintenance, input volumes, query rates and communication costs.
These will vary from file to file, and a goal of the COINS I is to
identify when and for which files duplication may be desirable. An
additional complexity arises from the essentially unpredictable require-
ments for query of files. It is possible that the servicing of one
Agency's requests on another Agency's system may swamp that system. The
integrated planning of systems which must meet requirements from all
Agencies will require experience and greatly increased cooperation.
b. Types .of Duplication
(1) File - All or part of a file may be dul)licated_ _Fn-" Avory 'l-
(c) In some instances a machine file of one Agency may con-
tain records derived, from many different sources. For example, Agency A
may enter order-of-battle information derived from the reports of Agency B.
When Agency B interrogates Agency A's order-of-battle file, how does it
know that information received is not that which it originated? There
is a danger that Agency B may treat the answer derived, from Agency A's
file as confirmation when, in reality, it is receiving its own information.
On the other hand., an Agency's file may contain conclusions based on
information which has been derived from sensitive compartmented activities
not available to the analysts in another Agency, or from information not
exploited by another Agency. However, if the file represents the Agency's
finished intelligence there may be honest differences of opinion. A
goal of COINS will be to identify those areas where division of respon-
sibility in the-establishment and maintenance of files is practical and
economical.
(2) Elements of Information Within a File - Various elements of
information within a record of a particular file may come from different
sources. For instance, Agency A has a military personality file. Though
the name and. location of a particular-individual carried in one of the
records was derived by Agency A, his rank and organization were derived
by Agency B. How does an analyst know which elements of the record were
derived by Agency A and which were from Agency B?
Again there `is a
SECRET
Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
danger that an analyst might consider the rank and. organization in Agency
B's file as confirmation of the information in Agency A's file. It is
therefore desirable that a file include the source (i.e., Agency or
report title) date entered and validity of information contained in the
file.
13
Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
INTERROGATION PROCEDURE
The principal objective of COINS is to improve the interchange of
information between intelligence Agencies by permitting remote consoles
in any participating Agency to interrogate the selected files of other
Agencies via a secure computer-to-computer link.
2. RESPONSE TIME REQUIRED
A design goal of COINS I is to provide a fifteen minute response
to all interrogations. This system is not designed to provide users
with either an instant response or with the capability to browse through
a file on-line. A fifteen minute delay for answers to interrogations
is a reasonable delay which is acceptable to all participating Agencies
for COINS I.
3. EXPECTED VOLUMES
It is impossible even to estimate the number of inter-agency
interrogations per day that COINS I must accommodate. In fact, one of
the early objectives of COINS I will be to dc;bermine the amount of inter-
agency activity desired. It is safe to assume however, that the majority
of interrogations will occur during the day shift of normal work days.
As soon as COINS I 'becomes operational the volume of inter-agency
interrogations can be expected to rise markedly as; (a) thenumber and
variety of files are increased, (b) the number of remote consoles in
each Agency are increased and (c) the users in each Agency become familiar
with. the system particularly with its capabilities and limitations.
4. INTERROGATION PRIORITIES
Each Agency has its own system of interrogation priorities and,
as indicated. below, there is not always a one-for-one correspondence.
a. Whenever DIA or State is interrogating an NSA or NPIC file they
will use the common priority system employed by these Agencies. Table I
illustrates the relative ranking of these two priority systems.
b. Whenever users in NSA or NPIC are interrogating the files of DIA
they will use their own interrogation priority system. The NSA RYE/TIPS or
NPIC system will examine the priority applied by the NSA or NPIC user and
automatically translate it to the appropriate DIA priority and insert the
appropriate priority lock according to conversion Table II.
x+14
. SECRET
Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
TABLE I
INTERROGATION PRIORITY CONVERSION TABLE
STATEIDIA TO NSA/NPIC
DIA NSA/NPIC
PRIORITIES PRIORITIES
R (ROUTINE)
P (PRIORITY)
U (URGENT)
------------------------------------------------------------------------
TABLE II
INTERROGATION PRIORITY CONVERSION TABLE
NPIC/NSA TO DIA
NSA/NPIC DIA
PRIORITIES PRIORITIES - PRIORITY LOCKS*
BLANK, 0, 1, 2 R (ROUTINE) 7?
3-4 P (PRIORITY) ??
5 U (URGENT) ??
*Fixed Priority locks will be determined upon approval of this plan.
15
y-' ,~-~
SF'l .P VT
Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
P.. .& --A A-d 1
5. SCOPE OF INTERROGATIONS
a. The most elementary information retrieval system conceived would
permit remote consoles in one Agency to interrogate (or look up information)
in another Agency's file. However, in such an elementary stage there
may be no provision for melding together into a composite answer
information obtained from several different files. Adoption of this simple
approach negates, to a large extent, the need for standardization of data
elements since the information is not to be merged for machine processing.
(1) This approach simply automates the situation which exists
today where each Agency periodically receives the reports published by
other Agencies. Then analysts search through and cross-reference these
reports manually in an effort to find the desired information. This is
a laborious and time-consuming process. Generally, these reports go
through a publication and distribution process and as a result the
information contained in them is quite old.
(2) Even so, there are some advantages to be gained in adopting
this procedure, particularly in the early development stages of COINS
when all organizations are learning.
(a) Information obtained from interrogating files in such
a system would be more current than that contained in a hard copy
machine listing or report, since such reports have generally gone through
a publication and distribution process. Such a system would not eliminate
the need for publishing or distributing hard-copy reports immediately,
but it should eventually reduce the number of copies desired.and perhaps
the frequency-of publication.
(b) Only the actual information specified in the interroga-
tions would be received.
(c) This approach eliminates the necessity of searching large,
bulky, and out-of-date machine listings manually, though'the answers
derived from interrogating several files independently would still have
to be melded together manually in the same fashion as they are today.
(d) This approach eliminates the necessity for standardizing
data elements and items, a subject which is quite parochial in some Agencies.
b. A more advanced information retrieval system, the goal for COINS II,
would permit one Agency to interrogate all of the appropriate files in COINS
with one interrogation. The individual answers received from such a multi-
file interrogation would be melded together, processed and formatted before
bei. printed out at the appropriate remote console. This approach has
all of the advantages of the elementary system, but eliminates the need for
manually cross-referencing the answers received from several unique
interrogations. However, this approach necessitates:
(1) Standardization of data elements
(2) Adoption of acceptable 'translations tables when data elements
cannot be standardized.
16
SECRET
Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
-3.~u~li 1_,~
INTERROGATION STRATEGY I
a. The method of processing incoming interrogations including the
search method employed by each Agency may be different as it is dependent
to a large extent on the configuration of hardware and software available
at each Agency.
b. There are basically two distinct types of interrogation strategies,
file-oriented. interrogations and subject-oriented interrogations. Each
appears to have its own set of advantages and disadvantages and both
should eventually be provided for in COINS.
(1) File-Oriented Interrogations
This is the interrogation strategy that will be employed
in COINS I. The user or analyst making the interrogation must have
knowledge of the file or files being interrogated. This requires the
exchange of much data between Agencies as well as-considerable study on
the part of the users. The user must know the:
(a) Name of the file(s) being interrogated
(b) Name of the field.(s) in the file(s) being interrogated
(c) Data elements (i.e., how the information is recorded)
in the file being interrogated.
(2) Subject-Oriented. Interrogations
This type of interrogation is one of the goals for COINS II.
In this situation the user does not have to know anything about the file(s)
which must be interrogated.(or searched) to derive the desired answer.
Simply stated, he merely indicates the subject on which information is
desired and, leaves it up to the system to find the necessary information.
It is fully recognized that the system itself must contain and respond to
all the rules and. discipline of file orientation. When using this type
interrogation, the user must provide as much amplifying information as
possible with respect to the elements of search (i.e., data items) to narrow
down the number of files to be searched. For example, if an analyst were
to submit an interrogation for information on the "3rd Infantry Regiment",
it would be necessary to specify the nationality. Otherwise, the'system
would. search all files in the system having information on military
organizations.
7. ASSUMPTIONS
It is assumed that:
a. Each member Agency will have detailed knowledge of the COINS
files of the other Agencies, particularly with respect to file names,
field names, content data elements, etc.
17
C~C'D 1P T
Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
to i t r 3 Z
b. Some stanrd.s will be established with repect to file names,
field names and data elements. This is essential if information is to
be extracted from the files of more thar_ one Agency and merged in
subsequent processing.
c. Each member Agency will keep other using Agencies informed of
any significant changes in file format, content or accuracy.
8. TYPE OF INTERROGATION
There are two distinct types of interrogations which must be accom-
modated in the system, specific (canned) interrogations and interrogations
making use of an inquiry language. Each type of interrogation has its
own set of advantages and disadvantages (See Table III).
9. SPECIFIC (CANNED) INTERROGATIONS
a. The named interrogation program contains and specifies the:
(1) File(s) to be searched.
(2) Field.(s) within these file(s) to be searched.
(3) Processing (i.e., mergining or summarization)-to be performed
on the desired records extracted and
(1i) Output format.
b. The answers (or results) to specific interrogations will be
automatically forwarded to the computer system of the requesting Agency
for:
(1) Output through its own remote system
(2) Input to another program for subsequent processing (i.e.,
merging with information received from another program, reformatting,
etc.)
c. Each Agency will make available to the community, information
relative to its existing "specific (canned) interrogation programs" for
those files which are to be included in COINS. If an existing specific
interrogation program does not satisfy the needs of another. Agency, the
following options will be made available to the requesting Agency:
(1) Write a specific interrogation program for inclusion in
the program library of the retrieval. system of the Agency having the
desired file(s). The program must be written in accordance with the
procedures and conventions established for that system. For example,,
SECRET
Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
TABLE III
ADVANTAGES AND DISADVANTAGES OF DIFFERENT
INTERROGATION TECHNIQUES
INTERROGATION ADVANTAGES
1. Specific or "Canned" a. Users do not need to be trained
Interrogation in the use of a retrieval language
Programs for a complex input-interrogation
format.
b. The use of short interrogation
statement reduces the number of:
(1) Key strokes required by the
users.
(2) Error-likely situation that
might occur when the users are
required to type in a long, rigidly
formatted interrogation statement.
2. Interrogation a. Permits a user to formulate &
Language enter immediately interrogations
which have not been pre-programmed.
This is particularly useful if
time is critical.
b. Eliminates the necessity of
writing a separate program each
time a new combination of info is
required by the users.
19
DISADVANTAGES
a. Means a separate program must be written
for each interrogation but the amount of
original programming required will be re-
duced through the use of common sub-routines.
b. Restricts the users to a canned set of
question & answers. Therefore, to obtain
the desired. info the user may have to:
(1) Ask more than one canned question
(2) Write a new program. The writing of a
new program will require time; how long will
partly depend on how much prior coding from
other interrogation sub-routines can be in-
corporated in the new interrogation program.
U)
L
developed which is common to all files in the
system, and. all interrogation & file mainte-
nance inputs.
b. A compiler must be developed & written to
translate each interrogation into a program.
This requires a large number of systems pro-
grammers who are not currently available.
c. Users are required to be trained in the
use of a retrieval language and a complex
input-interrogation format.
Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
`.'NSA might have to write a specific interrogation program for inclusion
in the DIA/IBM 14.10 system, and conversely, DIA might-have to write a
specific interrogation program for inclusions in NSA-TIPS. This means
that DIA would require a limited number of RYE/TIPS programmers and
NSA would require some IBM 1110 ISIC -proarammers. In October 1965,
arrangements were made for to join the UNIVAC-11.90
(2) Write a specific program for their own system to merge and/or
reformat the results (answers) received from other computer systems.
When the desired results (answers) are not produced from the data as
received from other computer systems, they might be obtained by reformat-
ting the data or by combining all or parts of data from two or more
specific interrogation programs. For example:
(a) An analyst at DIA can enter a specific interrogation
called "ABLE" from his remote console.
(b) The. DIA computer will recognize that in order to answer
this interrogation, it must in turn automatically initiate two specific
interrogations available in the NSA computer called "BAKER" and "CHARLIE."
(c) When the results of these two interrogations.are
received, the DIA computer will merge and/or reformat the results and
output them to the requesting remote console. When this technique is
used, the DIA analyst does not have to know the names of the specific
programs in the NSA system, but the DIA computer must have this informa-
tion.
(3) It should be noted that the DIA analyst could have requested
the results of NSAts specific interrogation programs by merely inserting
BAKER and CHARLIE. However, it would have required two separate inter-
rogations and. the results would not have been combined.
30. USER'S LANGUAGE
a. General
(1) Currently the "users language" in each system is different,
and., in each case, is dependent upon the sophistication of the software,
as well as the size of the individual computer complex.
(2) The long-range objective is to develop a computer-independent
userts language for the Intelligence Community. The two languages used
in COINS I will provide the experience necessary for establishing the
specifications for this language.
(3) Analysts using this language will be able to extract'the
desired records from the appropriate file or files in the systems, of.
other Agencies, after which these records can be either:
SECRET
Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
iWe aZk AX%1 ' `
Agency.
(a) Forwarded in their complete form to the requesting
(b) Summarized in some prescribed. fashion, in which case
the summarized results will be forwarded to the requesting Agency.
(1i) Upon receipt of these results (i.e., either the set of
complete records or a summarized version), the computer of the requesting
Agency can:
(a) Transmit the-results to the appropriate remote station(s)
in its complex.
(b) Use the results as input to another program for further
processing (e.g., merging with information received from the files in
the systems of other Agencies, further summarization or formatting.)
For the purpose of this paper only, two levels of sophistication
have been identified within the users language, and each level is defined
below:
b. Levels of Sophistication
(1) Level I - Users can specify the (a) file(s) to be searched
(b) field(s) to be searched in the specified. files and (c) elements of
information to be searched for in the fields specified. Entire records
meeting this criteria will be extracted. from the file(s) indicated and
forwarded to the requesting Agency for further processing (e.g., merging,
sorting, summarizing and formatting). Selection of records is based.on
."IF, OR and AND" relationships.
(2) Level II - Users can specify:
(a) The file(s) to be searched by name.
(b) The field(s) to be searched by name.
(c) The elements of information to be searched for in
the specified field.. In addition, the user can qualify his interrogation
using the full range of Boolean expressions (e.g., and, or, if and only
if). Further, the user can specify the processing to be performed on
the extracted. record's (e.g.) merging, sorting, summarizing and formatting)
before they are transmitted to the requesting Agency.
c. COINS I Approach
For the purpose of COINS I, only the specific (or canned)
interrogation and Level I of the users language will be used.
QV_rl? P74
Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
d.0 COINS ii
In this phase provisions will be made to add interrogation levels
and a standard query exchange language.
SECRET
Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
 -JJV uJl1Jll2j 1 V%O
IV ti
COMMUNICATIONS
1. COMMUNICATIONS SWITCH
Implementation of COINS requires the establishment of a dedicated,
secure communications network to permit direct, on-line, linkage of the
various remote access computer systems and remote stations in the
Intelligence Community.
a. DIA will act as the communications switch in COINS I. (See net
diagram in Appendix I). In addition, DIA has agreed to:
(1) Procure and install the communications and crypto-equipment
necessary to operate the communications switch.
(2) Rent the data links required. for computer-to-computer communi-
cations between: (a) DIA and NSA (b) DIA and NPIC and (c) DIA and CIA.
Actually, DIA will pay for the communications lines up to the NSA, CIA
and NPIC buildings.
b. NSA/Tl will provide the communications engineering guidance and
assistance to DIA for the communications switch in COINS I.
c. CIA will be the communications store and forward switch in COINS II.
(See net diagram in Appendix II). This phase will start in FY-68 with the
installation and. checkout of time sharing equipment for both the communi-
cations switch and a further update of the remote access computer system
at CIA. The store and forward switch in addition to its normal communicative
mission will be responsible for:
(1) Secondary security checking.
(2) Service and. equipment advisories and messa a on line and
equipment status for all directly interfaced equipment (i.e., non-availability
of service due to maintenance).
(3) Provide message receiving and error checking and recovery
software procedures.
(4) System accounting and. message servicing as determined to be
necessary for smooth and efficient operation of the system.
(5) Limited. conversion routines that may be necessary to provide
linkage of non-homogenious computer and. equipment.
2. POINTS OF AGREEMENT, COMMUNICATIONS
a. Transmission Rate - Initially the transmission rate will be 2+00
bits per second BPS with a gradual increase to 9600 BPS in COINS II.
However, it should be noted. that the HN9 is currently designed to handle
a maximum of 4800 BPS and therefore a modification of the HN9 will be
necessary to meet the 9600 BPS requirement.
23
SVC':R W"Ir
Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80B01139A000100140007-1
Ii tid i
b. Security
(1) COINS I
(a) All transmission will be protected to TOP SECRET SI.
(b) All computer complexes including all remot t t
i
ons
will be located in an area protected to TOP SECRET SI, e s
(c) If any computer is to have multi-level fil
pp es or remote
stations specific a
approval for such use will be obtained in accordance with
DCID 613. (See Reference 1).
(2) COINS II
A multi-level all source communications and. terminal system
will be developed for this phase.
c. Standard Communications Code (ASCII) - The American Standard Code
for Information Interchange (ASCII) will be used for transmitting data
between the various remote access computer systems.
d. Length-of-Message - Messages exchanged between computers will always
be transmitted in blocks of 150 characters. A message to consist of the
necessary contiguous multiple blocks. If a message is shorter than 150
characters or if the last block is less than 150 characters it will be
padded out to 150 characters using a space (i.e., octal 0505 as the padding
character.
e. Message Transmission - Messages will be transmitted every 15 seconds
until a receipt is received, from the receiving Agency's computer.or for a
reasonable length of time.
f. Time of Operation - The computers will be linked together and
available for remote interrogation 24-hours per day, seven-days a week.
Error Detection and Correction
(1) The odd parity check-bit will be used in the ei
ht le
l ASCII
g
ve
code and it should. provide sufficient error detection for the initial phase
of COINS
Th
f
EDC
1.
ere
ore,
equipment will not be required initially for
the data links, but may be procured at a later time if operating experience
7.T9'Y.r Y1t c
(2) This is a problem which will be the subject of continuing
discussions until it is resolved. For DIA to meet the implementation date
of 1 December 1966, a decision in this area is required by 1 June 1966.
h. Transmission of Files - Total files will not be transmitted between
systems as this would defeat the purpose and objective of this effort. in
the case of NSA, a maximum of 1024, 320-character records will be forwarded
per single interrogation. Similar restrictions will probably be imposed
by DIA and CIA.
24
SECRET
Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
 Approved For Release 2007/10/19: CIA-RRDD.P80BB01139A000100140007-1
i. Communications and Cryptographic Equipment .s Each of the
participating Agencies will be expected to pay for its own communication
and crypto-equipment, including the necessary spares. NSA and NPIC will
provide a communication line terminal (CLT) in their multiplexer to
receive the DIA data link.
J. Type of Subscribers - There will be two types of subscribers in
COINS:
(1) Users who do not have machine files or computer systems, but
who require remote access to the information in COINS. This type of user
only requires a remote station and merely interrogates and receives
answers from the system. The National Indication Center (NIC) is an ex-
cellent example of this type of user. The State Department link is an
example of this type of operation in COINS I. For this type of subscriber,
the communication switch will serve as the requesting computer.
(2) Organizations who desire to enter the COINS network and
have files in a computerized information retrieval system.
3. MESSAGES
a. Type of Messages
(1) Following are the types of messages that will be exchanged
in COINS:
(a) Requests (interrogation or query)
(b) Receipts
(c) Responses
I Answers to interrogations including negative answers
2 Messages indicating the files are not available for
interrogation or that a specific interrogation is not authorized.
(d) Service Messages
(2) Examples of various types of messages are contained in the
four displays included in Appendix III:
(a) STATE/DIA interrogation cycle of NSA files using:
(1) Specific (or"canned") interrogations (Appendix ILIA)
(2) Users Language - Level I (Appendix IIIB)
(b) NPIC/NSA interrogation cycle of DIA files using:
(1) Specific or. ("canned") interrogations (Appendix IIIC)
(2) Users Language - Level I (Appendix IIID)
25
Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
b. Message Formats
ti
(1) All messages exchanged in the system will be in a format
acceptable to the computer system of the receiving Agency, As a result.
the receiving Agency can treat incoming interrogation messages from another
Agency's system in the same manner as if the interrogation had originated
at one of its own remote stations. This means that the. originating Agency
must either:
(a) Require its users to submit interrogations in the exact
format of the receiving Agency, thus eliminating the need for computer
reformatting. (This is the approach being used. by DIA).
(b) Write a computer program to translate and format the
outgoing interrogation into a form acceptable to the receiving Agency.
(This is the approach being used by NSA and. NPIC).
(2) The four displays included in Appendix IIIA thru IIID depect
at a glance the:
(a) Sequential flow of messages between NSA and DIA
(b) Format and content of these messages at each stage in
the sequential flow
(3) These displays are intended to provide sufficient details to
permit managers in each Agency to:
(a) Determine what computer programs must be written
(b) Determine what internal operating procedures' must be
(c) Estimate the cost in terms of dollars and man-months
required to write the necessary computer programs.
4. METHOD OF OPERATIONS FOR COINS I
a. Requests (or interrogations) which require information from files
in another Agency's system will be transmitted to the appropriate Agency
via secure data link. The computer of the requesting Agency will continue
to transmit a request (interrogation) at 15 second intervals until a receipt
number is received. for the request from the computer system of the other'
Agency. Thp next request (interrogation) will not he transmitted until
the receipt has been received by the requesting Agency for.the,preceding
interrogation. This does not mean that thy.-receiving Agency does not queue
this request after it has been receipted for.
b. The computer system receiving the request will:
(1) First, receipt for the request
(2) Second, validate the request by determining that:
(a) The specific or "canned" interrogation programs indicated
SIR1 F,T
Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
r.-' -.pL' ~~~ .
are available and that the requesting organization is authorized to use
the program(s).
the system.
(b) The file(s) to be searched are currently available in
(c) Requesting organization is authorized to search the
files specified.
(3) Third, respond with service message as required (e.g.,
"Interrogation not properly formatted," etc.)
(4+) Fourth, prepare and forward a proper response message to
the requesting Agency's computer system containing: (a) the desired data
or (h) an indication that no data is available, or (c) an indication that the re-
q,Liestix4g!,AZeiry isn't aaithorled. to use either the specific program or search
the file requested..
c. The computer system of the requesting Agency will be responsible
(1) Handling its own bookkeeping. This includes such elements of
information as:
(a) The receipt or job number assigned to each request
(interrogation) submitted from a remote station.
'(b) The date and time request was submitted from the remote
(c) The number of the remote station submitting the request
or which is to receive the answer.
(d) The name, organization and telephone number of individual
submitting the interrogation.
(2) Maintaining records with respect to requests, receipt and
responses transmitted between the various computers. This will include
such information as:
(a) The receipt or job number assigned by the external Agency's
computer. It will also relate this job number (receipt) to the job number
(receipt) previously assigned to the request from the remote station.
(b) The date and time messages (i.e., request, receipt and
response) were transmitted to and.. received from the external computer.
(3) Processing or manipulating the information received in response
to an interrogation., This will involve such processing as:
27
SRCRR T
Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
 Approved For Release 2007/10/19: CIA-RDP80BO1139A000100140007-1
AJ L/ tJi