RECORDS MANAGEMENT PROGRAM--POLICIES ISSUES ORGANIZATIONAL STRUCTURE

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP73-00402R000100090005-0
• Release Decision: RIPPUB
• Original Classification: S
• Document Page Count: 13
• Document Creation Date: December 19, 2016
• Document Release Date: December 14, 2006
• Sequence Number: 5
• Content Type: MF

100090005
Approved For Release 2006/12/16: CIA-R402~GFO
Issues
SUBJECT Records Management Program - Policies 13-
Organizational Structure
1. Paragraph of this memorandum contains recommendations
2. Before 1961 responsibility for the Records Management Program
was centralized in the Records Management Staff. In 1961 responsibility
for establishing, maintaining, and directing records programs was
assigned to the Deputy Directors and heads of independent offices.
Decentralization was not absolute, however, because the function of the
CIA Records Administration Officer was retained with a minimal central
staff. He is charged with responsibility for furnishing staff guidance,
assistance, and coordination of the Agency program, and for reviewing
and monitoring the decentralized programs. Neither the centralized
program that existed before 1961 nor the structure that has existed
since has fulfilled the fundamental requirement of controlling the
growth of inactive x record material maintained in the offices and
stored at the Records Center.
3. The reasons the centralized program didn't work before 1961
are still valid and it appears that no useful purpose would be ss served
by considering total centralization as a reasonable alternative now.
There are also several reasons the program as established after 1961
hasn't worked either: $a4
 Approved For Release 2006/12/16: CIA-RDP73-00402R000100090005-0
Program composed of decentralized programs and this is the begin-
ning of the ambiguity. It is not clear how one program is expected
to be constructed out of several. The ambiguity is intensified by
is ambiguous. It says there will be an Agency Records
the distribution of authority and responsibility.
I,01Z A1k
authority to establish, maintain, and controllrecords programs
he ds of
is delegated to the Deputy Directors and/independent offices and
none is reserved for the "Agency program". The Agency Records
Administration Officer (Chief, Records Administration Branch, SSS)
is made responsible for furnishing staff guidance, assistance,
and coordination of the "Agency program" but there is no require-
ment that these services be used. He is responsible for "re-
viewing and monitoring", but-there is no obligation to accept or
respond to his comments and recommendations. The regulation
seems to say that the Agency wants it both ways: decentralized
programs but centralized control. A part of the problem is that
there is no authority and no structure to make it wor'r either
way or in combination.
b. A second reason the program doesn't work effectively
now is that we operate a central storage facility but exercise
no control over what comes into it, how long it must be kept,
or when and if it may be destroyed. Each of the decentralized
authorities exercises$kas these controls and jealously guards
LL
 Approved For Release 2006/12/16: CIA-RDP73-00402R000100090005-0
right to do so. Theoretically, completely decentralized
programs would include provision.for separate storage facilities
for each program. Clearly this is too outrageous to consider
seriously but it does suggest that perhaps some central control
should be exercised over the use of the Records Center, partic-
ularly if the managers of the Records Center are to be held
accountable for volumes retained, and charged with responsibility
for reducing them to a proportion which will fit within the
present space.
c. A third reason the Records Program doesn't work is that
storage is the only element of the Records Programs that attracts
only gets attention when the avail-
able space is used up. Other elements of the Program do not get
attention because there is almost never a crisis which can be
directly attributed to them. Managers like to have their records
programs remain unobtrusive. They gxx prefer to let subordinates
and
deal with problems of correspondence, reports, forms, records
ma:..gement
maintenance. There are no RH effective and systematic/systems to
deal with reports or forms in any of the records programs in the
AgencyC computer produced reports are literally transported
from computer centers on fork-lift trucks and external printing
of official forms costs the Agency more than a quarter of a million
dollars a year. We have no way of knowing how much is spent or
how much record material is produced for ultimate storage through
 Approved For Release 2006/12/16: CIA-RDP73-00402R000100090005-0
the uncontrolled development and printing of unofficial, bootleg
forms that are used in virtually every operation of the Agency.
We have no Agency correspondence program and no control over
copying machines yet these are probably the primary sources for
the creation of records that will ultimately require storage.
Moreover, there is no evidence that Agency management wants pro-
these
grams in tka vital elements of records management and control,
yet until we find some systematic way of dealing with the elements
of records creation we have no hope of dealing intelligently with
storage and disposal.
d. The Agency last issued reports and forms handbooks
more than ten years ago and the correspondence manual was last
published in 1955. They are out of date but the only
demand for revision comes from the Clerical Training Faculty
where they can be used as tools in teaching new clerical employees
some of the fundamentals. Efforts to revise these publications
in recent years have repeatedly bogged down in the coordination
process because it has proven impossible to get agreement among
the various echelons of management on the basic principles of
uniformity and standardization. Reports are generated by subor-
dinates to give the managern what they think he wants, not to
give him what he says he really needs. Management attitudes
toward correspondence are very largely esthetic and practices
followed in different components are usually developed in terms
of what is appealing to the eye of the senior manager or one of
his closely associated subordinates.
i~L'uJCI Yl
 Approved For Release 2006/12/16: CIA-RDP73-00402R000100090005-0
gained from uniform practices and the application of professional
so
standards are not/impressive in the individual organizational
unit that they become effective selling points in attempting to
particular f f =i,-i
persuade the manager that a P1M7.i2 Zt1 practice should be adopted.
The fact that a bloc format justified flush left (the Federal
Government standard) will save several seconds of a typist's time
by eliminating tabular indentation, and the fact that a particular
kind of file folder costs two instead of thirty cents and can save
one hundred dollars worth of space in a safe are regarded as
statistical minutiae when presented to individual managers. In
the Agency aggregate, however, hundreds of man hours and thousands
of dollars might be saved daily by the adoption of a few
scientifically proven standards.
e. The Agency has been getting along without the benefits
of professional records management standards for a number of
years and can xxnxxn continue to get along without them if Agency
management wants it that way. If that is the decision, however,
we should not retain a centralzs staff with pseudo-Agency respon-.
sibilities assigned to it. We should be content to allow each
of the several diaxnE decentralized programs to deal with them in
their own way. The Records Administration Branch should be
cupL~ji \ i`._
 Approved For Release 2006/12/16: CIA-RDP73-00402R000100090005-0
relieved of all responsibility for an Agency Program and should
become the records management unit for the Support Directorate
with authority and responsibility prescribed accordingly. This
does not seem to be a reasonable alternative, however, because
it would not only perpetuate, it would intensify the problems
we k now have with records management.
3. The problems of the Records Program in the Agency wxx? will
not be overcome by decree. They can only be overcome by clear state-
ment of policy, objectives, and ixta intent, and the creation of a
realistic structure to make the program work. We need:
a. The xx strongest possible expression of support from
the highest levels of Agency management for axiaxa a totally
integrated AgxxxyHw Agency-wide Records Management Program in-
cluding all of the elements of creation, maintenance, and dis-
position.
b. To formalize the existing structure, which is composed
in positions around
of some sixty - sixty-five people xxxxxg the Agency charged with
records management responsibilities; to provide for the staffing
of these position with S fessionally competent pxxpia
personnel; and, ideally, to provide a career service mechanism
for the personnel management of these people and positions.
Q-1 T
 Approved For Release 2006/12/16: CIA-RDP73-00402R000100090005-0
c. A formalized, gx regulated system governing the
coordination and approval of all actions relating to any element
of the Records Program.
d. An authoritative monitoring and review system re-
porting to the proper level of management with a regulatory re-
quirement for response to recommendations, implementation of
recommendations, and follow-up action.
e. A systematic way of exercising xxihmxity authoritative
control over materials accepted into the Records Center for
storage and the length of time that they will be held.
4. The Records Administration Branch and the Records Management
Board are reasonable management instruments for the implementation of
the Program. A clear statement of Agency records management policy is
required together with a sharp definition of the authorities and
responsibilities of each issued inregulatory format.
._ Many xi parts of the records managemenfunction r"e closely
related to the information processing function of system analysis and
design. Competent records management officers are systems analysts
capable of dealing with the design aNA improved manual systems. This
is, in fact, what records management officers do in the process of
evaluating records systems in the offices. In principle the design of
a hard copy file is as integral a part of the overall information
C"
'Wet a?V;
 Approved For Release 2006/12/16: CIA-RDP73-00402R000100090005-0
processing system as is a computer file in systems which require
automation, t e analyst must be able to deal with .,accordingly.
Manual systems have input, output, processing and filing procedures
and the competent records management officer must be able to deal with
all of these parts of the systems. Modern records officers must also
be capable of recognizing the need and potential for automating systems
and for designing manual systems which will interface with them smoothly.
Conversely, systems analysts dealing with computer systems
should have an understanding of, and appreciation for, the skills of
records managers. Computer systems analysts must have an appreciation
of the technical requirements for forms design in order to take ad-
vantage of the most effective methods of preparing k? input and output.
They should at least be conscious of the basic principles of reports
management and records skga storage and maintainence since the automated
systems they develop are intended primarily to produce reports which
be controlled, managed, and eventually stored in accordance with
sound records management principles.
/". The records management function is also closely related to,
but should be separated from, the archives function. The records
created, processed, maintained and stored today become the archives of
tomorrow. Careful and prudent selection of archival material is
facilitated by and dependent upon a well designed and effectively
managed records program. Early and systematic identification of record
 Approved For Release 2006/12/16: CIA-RDP73-00402R000100090005-0
material to be retained for archival preservation in a manner which
does not negate the requirement that records storage be minimized
requires that the records and archives program be compatible and care-
fully integrated even though they are separate functions.
Archives are the tools of the historians, who are students
and scholars. Occasionally the less xunsainx conscientious among the
historians may be inclined to dispose of a document after its substance
has been used and recorded in whatever historical account is being
written. This is a superficial attitude and not a prevalent one. It
is mentioned only to suggest that while there are significant areas of
interest common to the historical, archival, and records management
functions, there can also be some areas of basic conflict of interest
having
of the kind which can be most effectively dealt with by/all of these
functions under a single management.
9. In the critical review I submitted ?ss late in January I said
that it is not tovearly to hE? begin considering the ultimate disposition
of the SIPS Task Force. The notion that DDS management us must have
full responsibility for DDS systems regardless of whether they use com-
puters, desk calculators, or quill pens is just as valid now as it was
the
when we began/systems studies which have grown into the SIPS project.
We set out deliberately to develop the skills necessary to become
self sufficient, We acknowledged from the outset that we would have
to rely on the Office of Computer Services for technical support in
 Approved For Release 2006/12/16: CIA-RDP73-00402R000100090005-0
the skills of xxmf computer systems design and programming in order to
get the scope of our information processing system requirements
operational in some reasonable time frame without waiting to develop
these specialized skills for ourselves. Experience in recent months
has demonstrated that system responsibility for on-going applications
is just as intense as it is for new system development. The formation
of the SIPS Task Force acknowledges that skills in computer systems
design and programming and analytical skills representing knowledge
of the subject matter and functions for which systems are being designed
and operated must be responsive to a single management structure.
Maintenance of on-going systems, the development of new ones, the adju-
dication of priorities among them, and the allocation of available
resources to meet all of these demands can only be reasonably managed
within a single management chain. This says that we should not be
looking forward to the Ls lu-s+ rr of the SIPS Task Force, but should
be planning for its absorption into the Support Directorate.
10. Absorbing the task force into the Support Directorate
immediately raises the question about the location of the hardware.
We have said from the beginning that we see no need for the Support
Directorate to have its own hardware provided that the Office of Com-
ade uate
puter Services continues to be able to furnish kgkh support. Whether
computer systems designers and programmers have to be under the same
management structure as the hardware i/ a debatable issue, but there is
a great deal of argument to support the premise that they do not.
 Approved For Release 2006/12/16: CIA-RDP73-00402R000100090005-0
Computer service bureaus operate all over the country where centrally
located hardware services a large number and wide variety of computer
government
applications. Many/organizations have hardware facilities to serve
customers in other organizations. We have used some of these mxisa?
ourselves for back-up from time to time and of course, parts of our
payroll application has been run on hardware at the Treasury Department
for several years. It doesn't seem too unreasonable to suggest that
the Operations Division could function as a service bureau in the same
way. In a sense it operates that way now. Systems designers and pro-
grammers in the Management Support Division, now a part of the SIPS
Task Force, and other divisions of OCS are required to submit requests
for service almost as though they were part of another organization
weight
entirely. In any event the w xt of the arguments supports much more
conclusively that the analysts, designers and programmers should be
responsive to the manager who has the problem to be solved and then
that they should respond to the manager who controls the hardware.
While probably it is not prudent to pursue this now I believe our plan-
ning for the future should anticipate that it will occur he question
of whether or not the Support Directorate should have its own hardware,
however, need not be addressed now. If our applications, when imple-
mented, require some stand-alone configuration of computing equipment
it may make sense to have the whole thing in the DDS structure. Whether
this occurs or not, I don't see that it needs to affect our planning
now.
 Approved For Release 2006/12/16: CIA-RDP73-00402R000100090005-0
11. The reason for mentioning the information processing
the
organization in this paper is not only to identify relationships and
similarities with records management but to lead up to the suggestion
may
that you/want eventually to consider the establishment of an Assistant
xapty Deputy Director for Support for Information Processingy who would
have responsibility for the Support Directorate information processing
activities, records management activities, archives, and Agency history.
If that is a reasonable suggestion, actions you take now should be
consistent with that objective.
an
12. The long term objective of such/organizational structure
would be to provide single management direction to the interrelated
functions dealing with manual and automated information processing
systems; creation, maintenance, and disposition of their products; and
the identification and preservation of those products which have his-
torical and archival value. It should be at a level of the organization
which permits it to function across organizational lines and deal
adequately with systems which have an Agency-wide impact, to adjudicate
priorities, xx and allocate resources accordingly.
13. Acknowledging that the creation of such a position and com-
the question one
ponent is not a practical step to be taken now/becomes xxuiF of
selecting actions which will not be inconsistent with accomplishing
that objective later if you choose to go in that direction, and which
may be practical now. In an earlier paper the point has been made that
 Approved For Release 2006/12/16: CIA-RDP73-00402R000100090005-0
archives should the Records Program although the
archives function is closely related to the records management function.
The archives function is also closely related to the historical func-
tion. Each of these units should be brought together under a single
management. If this can be accomplished now, it should be but if it
isn't practical for any reason I would suggest that the archives be
established as a function separate from the Records Center and responsi-
bility for it to be assigned to the Historical Staff. The main reason
for suggesting this as a first step is to reinforce the notion that
archives are aapaa separate from records. Preferably, the archives
function would not be subordinated to the H historical function.
There could be an archives division and an historical division under
a single manager.
14. Another option might be to have a records division, an
archives division and a historical division under the single management
of a "Director of Documentation". If this combination of functions can
be accomplished in one action, it would be desirable to do so. If it
is not reasonable to accomplish this in a single action, the archives
and historical functions should be combined first with the records
function added later.
15. A third option would be to separate the archives from the
records function as a separate divisions under the same management.
The principle disadvantage of this arrangement as a first step is that