LETTER TO THE HONORABLE, RICHARD HELMS, DIRECTOR OF THE CENTRAL INTELLIGENCE AGENCY FROM PHILIP J. FARLEY

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP80B01495R000100020004-8
• Release Decision: RIPPUB
• Original Classification: S
• Document Page Count: 9
• Document Creation Date: December 20, 2016
• Document Release Date: May 11, 2007
• Sequence Number: 4
• Publication Date: July 22, 1970
• Content Type: PAPER

Approved For Release 2007/05/11 : CIA-RDP80B01495R000100020004-8
SECRET
r~~:llee
i-tats De. t. reier co
13 February 1969
FROM CIA COINS Subsystem Manager
SUBJECT: Objectives of the COINS Experiment
1. Design, develop and implement an experimental
system which will produce the necessa~y data to test
the feasibility and utility of an n eragency computer
based information system with on-line remote access
capability~to support the intelligence analyst) Follow-
ing are subobjectives which are implicitly derived from
the primary objective.
a. To isolate, efine and determine solutions
for a technics and securproblems
incident o designing, inwnsaTling, opera-
ting, using and maintaining an interagency
computer-based, on-line intelligence
retrieval system.
b. To provide practical training and actual
experience for intelligence analysts in
the use of an interagency on-line intel-
ligence retrieval system.
c. To ascertain the essenti l characteristic
of a file which make it suitable for in-
clusion in an interagen y on- ine intel-
ligence retrieval system.
d. To determine those elements of the total
system design which encourage and facilit
analyst usage and those cements whic
discourage or hinder usage.
."L To collect the necessary data on file
system fficiency .]~eli~ability and use
reactions opermil an eva a~E3$n of the
asp icy a itilIt o o erational
ementation of a I S-like sys em.
SECRET
Approved For Release 2007/05/11 : CIA-RDP80B01495R000100020.004-8
 Approved For Release 2007/05/11 : CIA -RDP80B015R000100020004-8
*440
KREI
2. Determine the a ements and ttribu e which
wo~k'd be critical to the ucceesss, of .a-_Tull system
-a" P
fo ::ontinuing community operations, #t evelo
ob'--tives andspecifications for an opera Iona system.
SECRET
Approved For Release 2007/05/11 : CIA-RDP80BO1495RO0010002OQ04-8
 Approved For Releas 2007/05/11 : CIA-RDP80B01495R000100020004-8
Objectives
1, Design, develop, and implement a system for the COINS
Experiment which will produce the nec scary data to test the
feasibility and permit an# evaluat'e"tie utility of an inter-
connected computer ace orma ion sys em with remote access
capability. A design objective will be to provide an average
response time of fifteen minutes or less to all interrogations.
2. Determine the elements and attributes which would be
critical to the success of a full system for continuing communit
operations and develop objectives and specifications for an
operational system. Some of the salient features that would
be included above would be network security, user training data
standardization,-a common network query language and criteria
for interagency files.
3. By means of the experiment, collect data to show
whether the coruunit*'s capability to exploit intelligence
information can be improved : -gym s+nZ-&.
By #. CL Y; `4
a.,,p4ore timely intelligence information available to
analysts in participating agency via remote access network of
-computers;
b. By obtaining more accurate information through
the resolution of data conflicts via high speed transmission
and manipulation capability.
4. By means of the experiment endeavor to collect data,
a. To show that better utilization o resources,
including time, men, and machines can be obtained through the
use,. of selected community files established and maintained in
one system;
b. To demonstrate the potential flexibility that can
be achieved in a COINS-type network such that the management
of intelligence production can be enhanced byhaving adequate
storage, manipulation, transmission and distribution of data
immediately available.
-~o 4 4
Approved For Release 2007/05/11 : CIA-RDP80B01495R000100020004-8
 25X1 Approved For Release 2007/05/11 : CIA-RDP80B01495R000100020004-8
Next 1 Page(s) In Document Denied
Approved For Release 2007/05/11 : CIA-RDP80BO1495R000100020004-8
 Ap,p For Release 2007/05/11: CIA-RDP80B01495R000100020004-8
v t SECRET '010,
10 February 1969
The meeting of the COINS committee was held on the 7th of
February at NPIC at 10:00 A.M.
no DIA representative present.
had the flu, so there was
was taking a physical so
there was no NIC representative present. However,
mentioned that the NIC will receive their equipment to put them on
COINS next week.
Test. COINS sub-system will query all of each others' files
during the week of 10-14 February. The attached paper describes the
)
test in detail.(.Sef - - C - t tI ) ( C ( $ )
Files. A rough draft of the complete identification of all
of the files on COINS will be sent out this week. However, the final
production will not be ready for quite a while thereafter. As an
example of what this file identification is to look like , copies of
passed around as an example. The example is attached. There was some
discussion on the security classification of this example. It appears
to be overclassified.
Contact for Assistance on Communications Task in COINS Master
I I
for assistance in meeting the deadlines for two of the tasks,
"communications" and "operational capabilities of network" for the
fiscal year 1969-Tasks in the COINS Master Plan.
He also discussed the various points made in the work statement
for the contract. The only software that the contractors will see is
A
Approved For Release 2007/05-r-U-RDP80B01495R000100020004-8
next described the contract which he hopes to let
 Approved For Release 2007/05/11 : CIA-RDP80B01495R000100020004-8
SECRET
Interface software. The proposal which the contractor gave him
mentioned security software, retrieval systems software, etc., bdt he
crossed all these out because the contractor will only have a SECRET
clearance. (See reference) George was concerned that there was
no DIA representative present because the DIA switch will be the key
item in this contract. This switch still gives him a great deal of
concern. George is taking this approach (that is, with contract
aid) because he has no talent pool available to help him meet the
deadline. He does have a second group composed of NSA talent, rather
than contractors, looking at what the ideal.:community system should be.
George felt that he does not get aselose to the Interface Panel
as he would like. It takes a long time to get their minutes but he
feels that they are changing; some of the little nitty gritty to
accommodate the various agency systems in COINS.
I recognized some truth in this, but felt that the task must be done
and only a contractor could stick to the job because he had no other
priorities. A second difficulty was caused by the fact that there are
some NSA planners of future communications laying out ways to study
their agency's communication in the future. COINS would be a natural
part of their study, so their efforts might be in paralled with the
contractor's efforts, and they would get in each other's way. He
already felt some antagonise from NSA people because they felt he was
using up all their time,
SECRET
Approved For Release 2007/05/11 : CIA-RDP80BO1495R000100020004-8
 Approved For Release 2007/05/11: CIA-RDP80B01495R000100020004-8
George added that contractors perhaps may not study the log
because of the clearance problem. NSA does have some open end
contractors and he might use them instead
II
much trouble. George has some objections to the use of contractors
per se, but feels in this case there is no other alternative.
Draft Charter on Security Panel. George felt that
group's production is going to be very long-ranged and broadly
based. He feels there is a need for a group to examine COINS
security aspects, of software in particular right now.
reported that the CIA member will be
and list of members of this security panel are in an attached document.
objected to the wording of paragraph c (recommendations)
are on the attached documents. The chief change is in Number 3.
Bob felt this panel would not have the time to write any software and
the recommendation was thus rewritten. All of the substitutions were
accepted and are more down to earth.
At this point we took a mental break before starting the CMP
objectives. There was considerable chatting on the CCC,
report on the CIA test on the CCC, about head of the line indexing in
CRS, retrieval systems in general, and the
Objectives.
system which has
expressed approval of the Support Staff's
objectives. (The third set of the attachment). He wished to substitute
these objectives to the ones in the COINS Master Plan. The NSA sub-
system manager disagreed; he liked the December CMP objectives and
25X1
25X1
Approved For Release 2007/05/11 : CIA-RDP80B01495ROO0100020004-8
 Approved For Release 2007/05/1 EJP8OBOl 495R000 100020004-8
stated that he would have a coordination problem within NSA if they
attempt to change their objectives. He thought that the community was
not in a position to change objectives now so soon before the test
and analysis period began.
I I
severe in objections to objectives 3 & 4 as succeeding correspondence
between agencies would indicate. He had put in the footnote to express
a different outlook on the objectives and regretted that they later
became such an issue. George said that if we list objectives we
should be able to understand what it is that we talk about. How do
we measure the objectives? What makes an objective a failure or
a success? Young also did not like the way that George had rewritten
the original PFIAB charge. He thought PFIAB has focussed on the
experiment alone in the 1965 paper. I Isays we are in a
Ae- one At
different ballgame thancAt-he PFIAB originally conceived. They wanted
outside agency terminals to the system in a simple
experiment. But for various reasons, now ha a complex tele-
communications system.
George gave Attachment 4 as his feeling as to what is the
most important COINS objective. He still believes in ultimate
goals rather than short-range: Young felt that CIA still doesn't
no
want to go that far. Charlie Stein, who or-had 0 chance to read
the objectives the meeting, said that he was still not quite sure,
there seems to be both intermediate and long-range objectives for
COINS. The first set concerns the technical feasibility, and the second
set customer satisfaction. But after you have proven the first, then
what? How can you prove the second?
S It' 1C), R Er
said that if
Approved For Release 2007/05/11 : CIA-R?P80BO1495ROO0100020004-8
replied that he did not feel as
 Approved For Release 2007/0 /11 -LA5R000100020004-8
NOW
5
NSA cannot accept the Support Staff's objectives, let's just use a
rewritten version of Objectives 1 and 2 in the COINS Master Plan.
There followed quite a long, almost philosophical, discussion on what
were the real objectives of COINS -- technical or people. NSA is still
worried about the technical difficulties and feels that they are not
settled yet.
Meeting was adjourned until next week when a meeting will
be held at NPIC at 1300 hours on the 14th (Friday). Subject will
be the continuation of the discussion on objectives. Each COINS
Sub-system manager will bring a paper which describes what the
considers the Number One objective and how it should be approached.
SO
Approved For Release 2007/05/11 : CIA-RDP80BO1495ROO0100020004-8