IDEAS ON PROCESSING INFORMATION
Document Type:
Collection:
Document Number (FOIA) /ESDN (CREST):
CIA-RDP80B01139A000200080004-0
Release Decision:
RIPPUB
Original Classification:
C
Document Page Count:
8
Document Creation Date:
December 15, 2016
Document Release Date:
March 3, 2004
Sequence Number:
4
Case Number:
Publication Date:
July 11, 1961
Content Type:
NOTES
File:
Attachment | Size |
---|---|
![]() | 468.59 KB |
Body:
Approved F Release 2004/03/17: CIA-RDP80BO14WA000200080004-0
C'-C-N-F -I-D-E-TN--T -l-A-L
CDDD3-D-82/7
11. July 1961
Ideas on Prooesain Iuforal4toa
1. In Cf3DLB-D-82/2 (Limited Distribution), I invited co atributians
to CODD3ss food for thought department.
2. Attached for information is such a contribution, a paper prepared
by Mr. Joseph Becker, Chief, CIA Automatic Data Processing .ff.
25X1
Approved For RLie`a 'P2 '(rJ/DVJi'' "I1TkB01139A000200080004-0
Approved For Relea -2004/03/17 : CIA-RDP80BOl139AOOW080004-0
C-O-N-F I-D-E-N-T--I-A-L
C
O
P
Y
29 June 1961
TO
FROM .
SUBJECT: Processing of Printed Intelligence Information
BACKGROUND
This paper is written in response to your desire to obtain some informal
views of whether we are going in the intelligence information processing field.
The characteristics of intelligence information processing are, except for
security restrictions, little different from those confronting the normal world
of Information and publication. Dipping back into an article written by. one of the
predecessor Chiefs of ODDR& E we find.that he had this to say about the infor-
mationsituation in the scientific community:
"There is a growing mountain of research. But there Is increased
evidence today that we are being bogged down as specialization extends . . .
The investigator is staggered by the findings and conclusions of thousands
of other workers .:. conclusions which be cannot find time to grasp, much
lose to remember as they appear . . . The difficulty seems to be not so
much that we publish unduly In view of the extent and variety of present-day
interests, but rather that publication has been extended far beyond our present
ability to make real use of the record. The summation of human experience
is being expanded at a prodigious rate and the means we use for threading
through the consequent maze to the momentarily important item is the same
as we used In the days of square rigged ships . ""
V%W
Were Dr. Bush writing about the status of publishing in the intelligence
community today he might have drafted an analogous paragraph as follows:
*Vannevar Bush, Atlantic Monthly, July 1945
25X1
Approved For Release 20(U ZF1 IbR@PfI01.JA9&000200080004-0
?a
Approved For Relea` '2004/03/17 : CIA-gDP80B01139AOOW080004-0
"There is a growing mountain of information being generated by our vast
USIB collection resources abroad. But there is increased evidence that
we are being bogged down by the multiplicity of separately directed
processing efforts ... printed data is piling up faster than we are able to
make use of it . . . The intelligence research analyst is staggered by the
volume of facts and findings reported to him in print . . . facts which he
cannot find time to grasp, much lose remember as they appear, nor from
.which he can deduce meaningful patterns for evaluation. The difficulty
seems to be not so much that we publish unduly in view of our world-wide
intelligence commitments and variety of community specializations, but'
rather that our rate of collection far outstrips our present ability to
process and analyze. Thus, no means exist for continually honing
requirements in a way that will ensure more precise field collection.
The summation of human experience in intelligence is being expended at.
a prodigious rate and the means we use for threading through the collected
maze of recorded information to find intelligence is the same as we used
in the days of Mata Hari."
PROBLEM
The problem in the intelligence community is that the size, cost, and
complexity of inter-agency information processing relationships are growing,
not at a constant rate, but at an ever increasing rate. Other influences, such
as the introduction of costly data processing equipment, serve to dramatize the
fact. that. there is no slackening of this trend. From purely an economic. viewpoint,
let alone others, it is imperative that the intelligence community devise some
mechanism for optimizing its programs in the area of information processing.
The CODIB group, and its predecessor organization, AHIIP, have vainly
sought solutions to the problem of interagency "compatibility" for the handling
and processing of printed information collected by intelligence in the field. While
we have achieved some measure of success through the introduction of minor
modifications to existing systems, no major blueprint or plan for total system
integration has been forthcoming. The reasons why this has not been possible
thus far are several:
a. Missions and interests of each member of the Intelligence
community differ;
b. Command lines responsible for Information processing activities are
independent and not always subordinate to intelligence control;
C-0- i-F-I-D-E-N-T-I-A-L
Approved For Release 2004/03/17 : CIA-RDP80B01139A000200080004-0
Approved For Releaa2004/03/17 : CIA-RDP80BO1139AOOW080004-0
C-O-N-F-1-D-E-N-T4I-A-L
c. The local or "in house" needs of any one agency usually have
priority over the needs of the other agencies;
d. A communal approach to the moor problems of intelligence
documentation cannot be followed up with executive leadership for action;
e. The value and benefits to be derived through Interagency compatibility
remain unclear because the notion of compatibility means different things
to different people.
DISCUSSION
As you requested, I spent some time with the various DOD service activities
concerned with information processing of printed data - Army, (John Kullgren),
Navy, W .W. Higgins and Dr. Yovite), as well as Air Force, (John Toler).
While the 3 services have identical objectives they are proceeding down
divergent data processing paths. Each has taken a different approach to the problem
and each employs different contractual assistance. Army's approach has been to
strengthen the Intelligence analyst's ability to collate printed data and Army is
using the Order of Battle function as a model; Air Force has emphasized document
retrieval through MINICARD and is seeking to provide the Intelligence analyst with
now. forms of data manipulation by computer; Navy has recently entered the. field
by trying a small experimental effort. Army employs RCA and uses a Sylvania
computer; Air Force Is associated with Eastman Kodak and uses IBM computer
equipment; and, Navy Ii soon to decide on its own avenue of exploration.
Historically, the information processing programs of the services had their
first beginnings in the "library" or "document file" activity. In post WWII days
the volume of Information collected abroad was comparatively small and the library
was able to function effectively along traditional lines. However, the virtual
explosion of collection media in the past decade has inundated the Intelligence
community with printed Information It is Incapable of processing. It Is this pressure
which has given rise to unconventional approaches and the expectation that automation
may possibly offer a fruitful solution to the problem.
Even though the 3 service Information processing programs are diverse, the
people who ran them are not unaware of the sameness of their objectives. They
freely exchange Information an each other's progress and share professional interest
In each other's work. Given present organizational structure and mission assign-
ment as a fact of life it Is difficult to precelve how the S service programs could be
made to mesh any further. While overlap does exist among the services it should
not be construed negatively as an outright case of duplication of effort. in reality,
Approved For Release 2jY 4 0 17F~fI~FP(~13 A000200080004-0
Approved For Release 2004/03/17 : CIA-RDP80B01139A000200080004-0
C-O-N-F I-D-E-N-T I?A-L/
iw~
each service is simply being responsive to the needs and commands of its own
environment. Moreovef, the entire field of information processing is so new and
experimental that multiple research programs tend to mutually reinforce one
another. Only through some reorganization can we expect to see radical system
Im~porovemeant or program consolidation.
During the last decade we have witnessed the expenditure of large sums of
money for R&D and for implementation of Intelligence collection systems without
a corresponding emphasis on the processing of and products realized. Inasmuch
as printed information Is the lifeblood of the national intelligence effort, we need to
introduce some means whereby we ensure the optimum use of it. Without such
change, the entire intelligence cycle is threatened with becoming unmanageable
and self-defeating - and, therefore, wasteful and unresponsive to its full purpose
and. responsibility. The processing of printed information should be viewed as
an independent program, deserving separate support in its own right, and placed
on a par with other major activities such as electronics, air defense, weapons
systems, etc
I do not believe that a prolonged study of the current situation, by eithor
in-house personnel or outside contractors, can add a new dimension to. our
understanding of the problem. CODIB discussions have succeeded in airing all
aspects of the need, of the underlying philosophy of compatibility, and of the
technical complexities that are involved. Listed below are some of the weaknesses
recognised by CODIB as existing in the multi-agency approach to the processing of
printed information:
a. Uncontrolled growth of information requiremeit sent to the field;
b. Subsequent wasteful use of collection resources;
c. Multiplicity of report formats and absence of standardization of field
preparation procedures for generating intelligence documents;
d. Very slow transmittal of documentary information from field to
headquarters;
e. Time-consuming, laborious, many-step dissemination mechanisms;
f. Duplication of headquarters analysis and indexing;
g. Independent growth of systems of storage and retrieval for the same
documentary material;
Approved For Release 2004/03/17 : CIA-RDP80B01139A000200080004-0
Approved For Releaa2004/03/17 : CIA-RDP80B01139AOOf080004-0
C-O-N-F-I-D-E-N-T-I-A-L
h. Poor Interagency oannaunIcation between similar repositories of
information;
1. Little or no user feedback;
J. Scattered research and development efforts;
k. Absence of rigorous statistical analysis of operations.
Everyone is therefore alrea y alerted to the problem areas and, in general,
agency representatives are personally disposed toward taking some form of
united action. The difficulty has been in deciding just what action to take .. .
not because the problem Is vague and requires definition but because all recognize
that centralized direction Is well-nigh Impossible under existing command structures.
As long as the specialized interests of the individual agencies and services prevail,
USIB members are to be expected to satisfy their primary Internal responsibilities
independently - and this is as it should be.
A logical alternative to present unilateral approaches would be the creation of
a USIB Information Processing Center. This Center would be responsible for
ensuring systematic preparation of printed information In the field, developing
rapid methods of transmittal, and performing technical services such as reproduction,
dissemination, microphotography, distribution, and machine language conversion.
The Individual services and other U8ID members would continue to operate task-
oriented EDP installations of their own to satisfy internal data manipulation requirements.
The Center would evolve its program along lines that would ensure the employment
of the most advanced forms of communication and electronic data processing
techniques. In essence, it would be responsible for doing the housekeeping data
processing jobs required by all of the USIB agencies, and for providing each USIR
member with input data in machine language as an end product. The recipients,
in turn, would tailor their Internal computer processing to satisfy user requirements
In their own environment. The Center would also have authority to conduct research
and engineering programs in support of its basic mission.
There Is considerable precedent for establishing such a joint effort. ELINT,
COMINT, and PI are prime examples of specialized information processing
organizations layered between the collectors and the users to perform common
support services. The proposed Center would be of the same type but its work
orientation would be toward all forms of printed intelligence (PRINT) information.
The CODIB recommendation to U81B represents a vehicle that could be used
to generate an NSCID for a National PRINT Center. Rather than studying the problem
Approved For Release 2004/03/17 : CIA-RDP80B01139A000200080004-0
C-O-N-F-I-D-E-N-T#I-A-L
Approved For Rele 2004/03/17: CIA-RDP80B01139A0(W0080004-0
C-O-N-F-I-D-E-N-T I-A-L
with Za people during the next year, more progress would be made if CODIB
drafted the NSCID and used the first allocation of USIB manpower to plan, organize,
and implement the Center.
SUMMARY
The volume of printed Information flowing In upon the Intelligence Community
has vastly increased. The members of this Community are now unable to exploit
fully these ending quantities of materials. Yet, these very materials are the
life-blood of the Intelligence production effort.
USIB members are aware of the growing problem they face in the information
processing field... and most have, Individually, undertaken to develop computer-
driven data processing capabilities in attempts to cope with the problem. Information
concerning these systems under development Is freely exchanged within the
community, but the programs themselves are largely unrelated to one another.
Given the present Community structure, this Is entirely natural.
CODIB has recently proposed a community wide survey of both the traditional
problems of: document and information control and the new problems and requirements
emerging as a result of the growth in member agencies of these various computer-
driven data processing systems. CODIB, however, and its predecessor organi-
zation, have already done much to illuminate the problems which plague all our
houses.
It appears to me that for fruitful study a new dimension is needed . . . a
new framework within which to study, plan, operate, and manage that vast
portion of the data processing problem which is common to all USIB members.
I do not feel the Community survey now under consideration by CODIB. provides
such a new dimension or framework. I believe a basic reorganization of our
present structure and system for the processing of printed intelligence (PRINT)
Is a pre-requisite to any major improvement In the Community's over-all data
processing effort.
I suggest, therefore, the creation of an USIB information processing center . .
funded, : managed, and staffed jointly by USIB members. This Center might be
called The National PRINT Center. The Center would coordinate the collection
and transmission of printed data from the field to assure technical suitability for
subsequent processing. It would perform common "housekeeping" services at
Headquarters for USIB members, such as storing, reproducing, Indexing,
disseminating, microfilming, and converting machine languages. The Center would
supply machine language Input to the various task-oriented EDP installations which
Approved For Release 92II6/W397I- I PT6IT1 1l '9A000200080004-0
Approved For Rel 2004/03/17: CIA-RDP80B01139A0 OQ 2 0080004-0
C-O-N-F-I-D-E-N-T-I-A-L
USIB members would continue to operate to satisfy parochial data mani tion
'requirements. TIC Center wadd also conduct R&D programs appropriate to its
.mission.
CODIA is the logical body to initiate steps to generate an NSCID calling for
the establishment of such a National PRINT Center.
25X1
C-O-N-F-I-D- E-N-T-I--A-L
Approved For Release 2004/03/17 : CIA-RDP80B01139A000200080004-0