DISCUSSION ISSUES AND WORKING GROUP EXPECTATIONS
Document Type:
Collection:
Document Number (FOIA) /ESDN (CREST):
CIA-RDP84B00890R000400060016-3
Release Decision:
RIPPUB
Original Classification:
S
Document Page Count:
44
Document Creation Date:
December 14, 2016
Document Release Date:
July 7, 2003
Sequence Number:
16
Case Number:
Publication Date:
October 5, 1981
Content Type:
MF
File:
Attachment | Size |
---|---|
![]() | 1.58 MB |
Body:
Approved For Release 200$$jI#T CIA-RDP84B00890RQ060Q6-3
IO la"
5 October 1981
MEMORANDUM FOR:
Members of Strategic Planning Working
Group on Information Handling Facilities
FROM:
SUBJECT:
Discussion Issues and Working Group
Expectations
ATTACHMENTS:
a.
b.
c.
d.
Paper titled "Overview of Information
Handling Facilities Concern"
Paper titled "Terminals"
Paper titled "Data Bases"
Paper titled "Data Transmission,
Dissemination and Distribution"
In preparation for the Working Group Meetings on
Strategic Planning for Information Handling Facilities,
the IHSA Staff has written four papers; one an over-
view and the other three on topical areas (see
attachments). These papers are intended to provide
information on current status, focus the issues,
and present key questions--with the expectation that
the answers will aid in setting planning goals and
objectives. The papers are being made available
to Working Group Members one week in advance of the
first meeting to help in preparing.
In following sections a summary of the overview
paper is provided, and for each of the topical areas,
the planning issues and expectations of the Working
Group in responding to the issues are presented.
First, however, it should be clear that the Working
Group need not be constrained to those issues to
follow; indeed, participants are encouraged to
present other issues they believe to be pertinent
to the planning process for Information Handling
Facilities.
25X1
Approved For Release 20RK, : CIA-RDP84B008
Approved For Release 2003108/1 , : CIA-RDP84B00890R000400060016-3
2.0 SUMMARY OF OVERVIEW PAPER
Growth rates in message traffic, VM service, and
disk storage are presented, with an explanation
that the growth has been absorbed by technology
advances. The paper establishes the linkage
between numbers of terminals and the need for
Data Processing and Communications Facilities,
and suggests that planning activities clearly
define this linkage.
While pointing out that technology has absorbed
much of the growth in Information Handling Facilities
notes of caution for the future are expressed.
Office Automation Services are singled out for
the potential of having a profound impact on
Information Handling Facilities during the planning
period. The paper points out that payoff assess-
ment will be increasingly important as the Agency
invests in Office Automation.
3.0 TERMINALS
The Issues
During the eighties the Agency will be challenged
by the need to provide a terminal for most manage-
ment, professional, and clerical employees. The
investment for just the terminals could exceed
$100,000,000 if we are to fully meet this need;
this presents a question of affordability. Accord-
ingly, we must understand better who is to have
terminals, when, and what type.
Working Group Expectations
o To provide estimates of numbers of terminals,
by type, to be installed each year during the
planning period (through 1989).
o Define characteristics of terminals in regard to:
o Data Processing
o Local Storage and Retrieval
o Graphics
o Electronic Mail
o Local Network Support
Approved For Release 200/LQ~YRT CIA-RDP84B00890R000400060016-3
Approved For Release 200ff~/ p CIA-RDP84B0089OR000400060016-3
o Develop planning assumptions for distributed
versus centralized functionality.
o Review current and planned programs to determine
deficiencies in satisfying Agency needs through
the eighties.
4.0 DATA BASE FACILITIES
The Issues
During the eighties we will be challenged by:
o Increasing quantities, sizes, and types of
data bases.
o Increasing numbers of users.
o The need to make data bases more widely
accessible.
o The need to provide easy to use tools
(Query Languages, statistical analysis,
modeling, graphics, etc.) to allow effective
utilization of large quantities of data.
Working Group Expectations
o Establish growth estimates for quantities
and sizes of data bases through the planning
period (1989). Characterize types of data
bases. Estimate the number of users for
each type of data base.
o Describe Data Base related tools that
will be needed for effective utilization.
o Provide estimates for needs to access
community data bases.
o Express strategic, or qualitative, arguments
which should be factored into the planning
process.
o Review current and planned programs to
determine deficiencies in satisfying Agency
needs through the eighties.
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B00890R000400060016-3
SECRET
5.0 DATA TRANSMISSION, DISSEMINATION AND DISTRIBUTION
During the eighties the Agency will be challenged
with providing the necessary communications and
dissemination facilities to carry and disseminate
increasing volumes of information to the right
people in a timely manner. Trends, which will
have an impact, include:
o New overseas facilities and procedures,
such as CRAFT.
o Dispersion of more terminals in the
Washington area and the need for connectivity
(terminal to terminal, terminal to
multiple computer systems).
o More information, particularly open
source information, will be distributed
in electronic form.
o New Office Automation functions such as
electronic mail, teleconferencing, and
shared electronic files place demands on
local communications capabilities.
There are several large programs either in develop-
ment, or in the planning stage, which provide
increased facilities for carrying and disseminating
information. There must be an assessment of the
adequacy of these programs to take us through
the eighties.
Working Group Expectations
o Characterize and provide estimates for new
operating procedures and new services which
drive requirements for distribution and
dissemination facilities.
o Express strategic, or qualitative, arguments
which should be factored into the planning
process.
o Review current and planned programs to deter-
mine deficiencies in satisfying Agency needs
through the eighties.
25X1
Approved For Release 200MMST CIA-RDP84B00890R000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Overview of the Information Handling Facilities Concern
1. The Dynamics of the Agency's IHS Environment
In spite of the fact that the personnel dedicated to IHS services in
the Agency have actually been declining, the volume and variety of services
have been steadily growing. In fact, the.growth has generally been
exponential over the past few years, and the compound percentage rates have
been high.
Figures 1 and 2 show the Agency message traffic carried by the Office
of Communications. As can be seen, the growth approximates exponential.
The correlations of the data with exponential regression curves are good,
and the growth rate is 10.8 percent per year for narrative traffic and 28.1
percent per year for data. These data reflect the urgent need for the new
technology embodied in the MERCURY project, because the capacity growth
capability of the current technology has been exhausted.
In the area of data processing, there has been a similar service
growth. Figure 3 shows that the number of logged users on ODP's VM service
has grown at a steadily exponential rate for the past four years. Since
the VM service supports real time transactions, its growth reflects that in
office automation functions. A similar story is told in the area of batch
processing. Figure 4 shows the daily computer time required for batch for
1980 and 1981; earlier data was not available. Even so, the data
correlates well with a regression curve reflecting an exponential growth of
18.5 percent per year.
A comparable history obtains for on-line storage. Figure 5 shows the
growth in such data, both in megabytes of on-line storage and in the
associated direct access storage devices (DASD). Although the data is much
less frequent--annual, rather than the monthly data available for VM--the
same exponential growth patterns is evident. In this case, the storage
growth rate is 20.2 percent per year: a value comparable to VM growth.
Because there are strong indications of greatly increased intent to use
IHS services, these growth rates probably have to be considered a base in
projecting what the demand might be in the near future. The recent spurt
in terminal acquisitions and requests, is detailed in the Terminals paper,
exceeds a 20 percent per year growth rate. There has, in fact, been what
might be considered an "explosion" in demand. Terminals, whether word
processors, data terminals, or special purpose units, are the system access
devices for the IHSs. It is reasonable to expect, therefore, that future
demands for communications and data processing services correlate strongly
with the number of terminals.
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
25X1 Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Next 4 Page(s) In Document Exempt
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
There is also a natural progression of interaction skills relevant to
users access of IHSs via terminals. The beginning applications are likely
to be word processing, cable processing, or simple file structures. Once
these have been mastered, there is a natural progression into more complex
functions. Users might then develop complex, limited file structures, or
write their own software to perform routine file loading functions or
correlate file data in special ways, or they might analyze data using
statistical packages. Ultimately, they might get into writing their own
data analysis programs. Thus, the IHS service requirement per terminal can
be expected to expand continuously, deriving the growth in service
requirements above that related' strictly to the number of terminals.
In considering the future, what we need, when we need it, and how we
are going to get it, the rate of technological innovation is important.
Almost all of the data processing productivity increases and a good deal of
that in communications has been provided by technological innovation in the
form of more performance per dollar. The chief engine for these advances
has, of course, been the integrated circuit. A prime example of this is
shown in figure 2. On-line user growth on VM has been provided by steadily
advancing mainframe performance; we have not had to change the
architecture. Solid state devices will continue to advance rapidly in
capability, but there is a question as to whether the past rate of advance
can be sustained. After a while, any technology runs up against physical
limitations. There is then relative pause until a new technology is
available to carry the functionality up to the next plateau.
An example of an area where technology has not saved us is direct
access storage devices. We now have almost 500 disk drives in the RCC and
SCC, and many more elsewhere. The large numbers of these peripheral
devices relative to the number of mainframes are starting to present all
kinds of problems, including reliability and operating software
interaction. That the problem is not worse is due to advancing storage
technology, as reflected in the difference between the 20.2 percent per
year on-line storage access and the 8.4 percent per year of DASD in figure
5.
The concern is that we become too reliant in our planning upon
technological innovation to solve all our problems, expecting and planning
for miraculously advancing performance/cost characteristics. There are
already some instances of the increasing difficulty of advances in solid
state technology, such as the disappointment that bubble memory technology
has provided to date. We need to be prudent in our planning, recognizing
that there will be technological advances and perhaps planning on them
where they are well indicated, but not depending upon them across the
board.
The providers of IHS services have been struggling to keep up with the
demand reflected in the foregoing data. Not all of the steadily increasing
productivity was provided by advances in communications and data processing
-7-
.SECRET
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3 -
technology, however. Unfortunately, because of resource constraints that
obtained, some of it was provided by drawing down on engineering and
development resources. That has left the providers in a somewhat weakened
position to respond to what may well be an accelerated rate of growth over
the historical levels. The implications with respect to development
resource requirements need to be considered in the process of projecting
future system capabilities.
2. Office Automation: A Current Planning Focus
An important factor in determining the IHS requirements in the
foreseeable future is office automation. It will be a large creator of
demand for data processing and communications services because the demand
will vary at least in proportion to the number of terminals. The number of
terminals, in turn, and particularly those connected to the central
systems, is going to increase at a very high rate over the next several
years--a rate limited by our ability to buy them, and install them, and
support them, not user demand. Moreover, the demand represented by each
terminal should also increase as users become more knowledgeable in taking
advantage of the functionalities available.
Included in the office automation category are such services as:
o word processing
o electronic mail
o cable preparation and sending
o information dissemination, distribution
and electronic filing
o small file construction and manipulation
o personal activities support
These are the services to be provided in SAFE and partially supplied now by
the RCC and SCC. The bulk of those office automation services usually
provided are supported in the VM environment. The advent of SAFE will
provide relief to the RCC VM but, even so, the needs of the other
directorates will keep the demand on VM increasing.
A principal concern in planning for this growth is "what is the
payoff"? What productivity contribution derives from the investments in
new office automation capabilities? The benefits fall into two general
categories of automation: tasks and functions. It is usually fairly
straightforward to assess the benefit of automating a task, since that
normally eliminates the need for human labor.
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
The value of automating functions is more difficult to assess. The
benefits of the latter are usually the provision of a superior capability
and the increase of professional productivity. Both are much softer things
to assess. As a practical consideration, also, it should be noted that the
creative or analytic process can only be expedited so much by automation.
There is a natural "gestation" process involved in the development of
problem formulations and conclusions; and that has its own time
requirements, unique to the application. The principal benefit of office
automation with respect to functions is the alleviation of detractions and
.distractions to the creative and analytic processes so that more time can
be devoted to them, rather than to expedite them.
In spite of the difficulties, we will need to assess the payoff of
office automation with respect to both tasks and functions in a
quantitative manner. It is needed to be able to deal with such questions
as: How great an investment is worthwhile for each type of individual?
How are these factors affected by the performance/cost character of the
equipment? For the near term future, we do not have to worry about
investing too much--we cannot provide enough capability fast enough to get
anywhere near a point of marginal payoff versus cost. In the 1985-1989
time period covered by this strategic planning effort, however, the
marginal investment concern is germaine.
It is not practical to think about a comprehensive, quantitative
productivity payoff assessment within the timeframe of the development of
this first strategic plan. That is something that will take concentrated
attention and will have to come in subsequent refinements. (The
measurement of office productivity for other than basic tasks is an
industry wide problem today. No commonly accepted methodologies are
known.) What needs to be done at this point is to develop some sort of
picture of office automation needs and priorities on the basis of a
qualitative sense of payoff. Generally speaking, the question is, what is
the priority order of services needed in both the task and functional
domains?
3. Some of the Top-Level Questions
This first strategic plan will be based primarily upon the definition
of functional requirements, their priorities and the schedules that derive
from the working groups. It is recognized, however, that these are
developed in the light of available technology and its relative cost.
Thus, perceptions of technological innovation are a dimension of the
planning effort. Answers to the following general questions should do much
to illuminate the time-phased performance goals for providers of IHS
services.
o What sort of professional functions should be
automated and with what priority?
-9-
SECRET
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
o How important is the automation of menial tasks,
reducing the need for low-skill labor, and
making the slots available for analytic
personnel? Is there a security motivation?
o How important is it to provide a powerful support
environment so that analysts can efficiently develop
their own small applications packages?
o What new technologies are seen as having
a high payoff in the office automation
environment, such as digitized voice or
teleconferencing?
o How will overseas stations differ from
headquarters in terms of required office
automation functions? Should the objectives
be to make overseas stations appear to users
like just another station on the net?
o Is the SAFE architecture perceived to be
appropriate to support office automation
functions elsewhere than NFAC? If so, where
and why?
o How important is it to introduce open source
material directly into the Agency distribution
system? Is there an adequate user community
to justify the likely investment?
o Is a foreign language capability needed within
the integrated system context? If so, what
is the relative priority and timing?
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Information Handling Facilities
The importance of the "terminal" and its associated
peripheral devices, such as light pens, disks, printers, etc.,
and its impact upon the Agency is most aptly stated in the Joint
Strategic Plan, 1981- 1987, of the Office of Communications and
Office of Data Processing:
The growth in number and importance of remote visual
display and printer stations presages a process which
will find almost every employee using a terminal as an
essential tool by the end of the decade.
It's the statement that followed this one that sums up the
purpose of this paper: "If this concept is to be realized,
planning and budgeting must take place now."
Terminals are identified as a specific dimension of the
strategic planning for two reasons: they comprise one of the
most costly components and they imply and produce demand for all
the other information handling services. They are also of
concern because the current demand seems to be explosive in
character.
What is a terminal? or is it a Cathode-Ray Tube (CRT)? or a
Visual Display Unit (VDU)? or a multi-function workstation? or a
console? or a stand-alone word processor? or an electronic
typewriter, etc.? Yet, this terminology problem reflects the
very nature of the problems that we encounter in discussing
terminals and their use in the Agency. For discussions at a very
general level, the vague notion of a "terminal" and what that
encompasses is often sufficient, as for example, in the preceding
paragraphs. However, at any more specific level of detail, it is
this very problem of defining what terminals are, and what they
can do, and what we want them to do, and how may of each kind
should we have, that poses many of the difficult issues that face
this Agency. Diagram 1 categorizes the many different facets of
how we view terminals in this Agency.
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
25X1 Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Diagram 1 reveals the complexity of the problems of defining
terminals because of the sometimes complicated relationships that
exist between what a terminal does, how it is configured and the
characteristics of a terminal. Out of the many possible
definitions that we could use for a terminal, there exists two
sets that are probably viewed as the two major categories that
terminals are placed in today. These two sets are:
1. Terminals that are purchased as stand-alone word
processors, NBI's, Laniers, CPT's, etc.
2 .'All the other terminals that are not in the first
category.
It is obvious that the distinction we make for purchasing
standalone word processors is fast fading, but given today's
environment, it is a sufficient enough distinction in order to at
least describe the quantities of terminals that exist in the
Agency. The quantities of Agency terminals in each of these
categories is given in Tables 1 and 2. These figures are
estimates and are used only to provide an approximation of the
terminals that exist today (September 20, 1981).. With the
current demand for terminals, this status is changing rapidly.
Table 1 identifies the stand-alone word processors that are
leased and/or purchased in the Agency by directorate. We are
currently dealing with 12 different vendors in the procurement of
these systems. They are: ARTEC, BURROUGHS, CPT, DEC, IBM,
LANIER, LEXITRON, NBI, QXY, SAVIN, VYDEC and XEROX. This myriad
of word processing systems is one reason ODP is currently
drafting an RFP to standardize on an Agency word processing
system.
Table 2 identifies the quantities of terminals that do not
fall within the first category. Again, these numbers are best
estimates provided only to give the reader an appreciation for
the approximate number of terminals we are dealing with. Table 2
only identifies the more popular brands of terminals in the
Agency. Special terminals, such as the Ramtek color graphic
terminals, and others that are few in number are not identified
in this table. The total number of terminals is estimated to be
approximately 1550. As pointed out, this number does include 87
Send/Receive 733's, that in some cases, are used only for the
purpose of a hardcopy device and not as a terminal.
-3-
SECRET
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
25X1 Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Next 1 Page(s) In Document Exempt
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
The total terminals identified in.these tables is 2,358 along
with 559 associated printers. These quantities are some measure
of the terminals as they exist today in the Agency, with the
exceptions noted.
On the following pages the GOALS and ISSUES revolving around
terminals is discussed.
Several major goals and issues need to be resolved by the
working group. The major goal, on which everything else is
predicated, is to determine the quantity of terminals that will
satisfy each of the directorate's requirements for each year
until the end of the decade. That this figure is growing so
rapidly is evident from the graphs depicted in diagrams 2 and 3.
The first graph, shown below in diagram 2, displays the
growth of stand-alone CRT word processing terminals within the
Directorate of Operations within the last five years. Similar
growth (maybe not as rapidly) is occuring throughout the Agency,
as our different offices learn of the utility and efficiency of
word processors in the office environment.
The second graph shows the likely annual acquisition rate of
terminals indicated in the first draft of the RFP for an Agency
word processor on the basis of user indications. It is
contrasted with the same data in the third draft three months
later.
The point being made by these two diagrams is that the
requirements for terminals is dynamically changing on a month to
month basis as more people realize the capabilities of these
systems to improve their office productivity and the quality of
their product. A realistic goal of the number of terminals we
are dealing with needs to be made so that appropriate planning
and budgeting can take place.
A modified view of the original assumption that was made in
the OC/ODP Joint Strategic Plan is presented in Table 3. Let's
assume that for each directorate, one could arrive at a perceived
ratio of terminals for each of the following three categories of
personnel, Professional, Clerical and Technical, in each of the
Directorates, re given for personnel in Headquarters and
SECRET
-6-
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
25X1 Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Next 2 Page(s) In Document Exempt
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Reasons for not choosing 100 percent ratios in each column
are based upon the facts that there would be some positions for
which there would be no need and some positions which would use a
shared work station. These ratios are rough estimates taken to
initiate this assessment process, and should be reestimated by
working group members. If they are correct, on average, there
exists a requirement to procure in the neighborhood of 10,000
terminals by the end of the decade.
If 1988 is assumed to be the target date, the growth curve
for installed terminals will look something like the curve in
diagram 4. Given that over 50 percent of our current base of
terminals will need to be replaced, the Agency must then acquire
terminals at least at the rate of approximately 1250 per year.
Today's terminal costs in the neighborhood of $5,000 to
$14,000 for the hardware, including the cost of a hardcopy device
for every two terminals. Installation charges can range from
several hundred dollars to install a terminal on the RED GRID
system to a couple of thousand dollars to install a terminal in
the Wideband Bus System. $10,000 is probably pretty close to an
average, in terms of today's technology. At that price, the
total cost to provide those 10,000 terminals is $100,000,000 in
81 dollars.
Leasing stand-alone word processors, as we are doing now, is
not an attractive alternative. In fact, the high cost of leasing
is a strong motivator to purchase. Today, the leasing cost for
the approximately 600 stand-alone word processing terminals in
the Agency is approximately $250,000 a month, or $3,000,000 a
year. In general, our total unit lease costs equal the purchase
price at somewhere between two and three years.
It seems reasonably clear, that if the number of terminals
required is similar to that projected, the direct concern is the
strategy of acquiring them: how to get a lower average cost of
an installed terminal and what the time-phased acquisition
program should look like. This latter then needs to be further
defined in terms of its composition in terms of major categories.
To get a lower average cost there is probably only one option
and that is to plan for a price break deliverable by commercial
technology, implemented in accordance with TEMPEST requirements.
-10-
SECRET
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
25X1 Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
That means that a time frame, such as 1983 to 1984 should be
defined, during which new commercial technology will be
exploited. Further questions are then, for what types of
terminals or terminal functionalities? These expressions have to
be broadly stated, since future terminal capabilities cannot be
projected with a high degree of accuracy.
Probably the big question is, how much are terminals
worth to the, Agency in FY-81 dollars? This can a translated as
saying that the terminal question may be more a design to cost
problem than a question of how much performance is
desired/required.
IV. QUESTIONS
The questions that need to be addressed in dealing with
strategic planning with respect to terminals includes:
1. What should the ultimate percentage allocation of
terminals be (update of Table 3) and when should it be
achieved?
2. What are the characteristics of the quantity of
terminals that have been determined in response to
question 1?
a) What are the requirements for word,processing?
b) What are the requirements for data processing?
c) What are the requirements for office automation?
Electronic Mail? Time Management? MIS?
d) What are the graphics requirements?
e) What are the availability requirements with
regard to both the terminal and system?
(distributed vs. centralized functionalities)
3. What are annual terminal requirements by directorate by
type of terminal?
4. What are the peripheral requirements for terminals?
What are the requirements for local printers? local
communications?
-12-
SECRET
25X1
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Information Handling Facilities
Strategic planning activities for data base facilities are to
consider the needs for creating, accessing, and processing text
and non-text (formatted, graphics, imagery) information.
We believe that the requirements for automated data bases will
expand at a rapid rate during the eighties:
We will be challenged by
o Increasing quantities and sizes of data bases
o Increasing numbers of users
o The need to make data bases more widely accessible
o The need to provide easy to use tools
(Query Languages, statistical analysis,
modeling, graphics support, and perhaps
generalized program libraries) to allow
effective utilization of large quantities
of data
o The need to understand new data base
technologies and how to effectively integrate
those technologies into the Agency environment
Planning groups are tasked to reduce this generally described
challenge to more specific goals and objectives so that Infor
mation Handling Programs can be initiated and targeted. In
doing so, we will review the status, and planned enhancements,
of current facilities and identify any deficiencies; we will
extract data from the current environment which determines
trends from which we can project; and we will ask user
communities to respond to issues and questions, the resolution
of which is intended to lead to specific strategic user needs
for the eighties.
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Data Bases within the Agency are diverse in size, use, and
complexity. They range from those that are small, run on
minicomputers for a very limited number of users, to those that
are large and complex which reside in large hardware systems with
many (and somewhat diverse) users. In general, the Agency's Data
Bases have been created and maintained for a specific component
and for a specific purpose. There has been relatively little
effort in examining commonality-'of use between components. This
is understandable due to the project and mission orientation of
many Agency components.
Large and complex Data Bases usually reside in (on) the large
computer systems in the two ODP computer centers, the Ruffing
Center or the Special Center. This is of necessity because of the
large amounts of information to be stored and the attendant effort
necessary to "run" and "maintain" them.
The DO and COMIREX Automated Management System (CAMS) Data Bases
reside in the Special Center because of security and isolation
considerations. These Data Bases are custom tailored to specific
needs and range from large document DBs to operational support and
tasking DBs. Included are hardcopy with index DBs (DORIC/W-
ALLSTAR) and many records DBs.
NPIC maintains the Installation Data File and related information
within the NPIC Data System on Univac hardware
This Data Base is used for NPIC support and operational support to
the community via COINS.
The remainder of the Agency is provided Data Base support in ODP's
Ruffing Center. Again, these are generally tailored for specific
use and range in size and complexity. Access and use is largely
dictated by user (customer) requirements.
The Agency has attempted to provide Data Base services by
providing a "standard" Data Base Management System for information
storage and retrieval. Currently, the Generalized Information
Management (GIM) System is provided for specific relatively large
applications and RAMIS is provided for smaller, less sophisticated
Data Bases. RAMIS is used extensively for its Report Writer
capability.
-2-
SECRET
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
25X1
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
The current provision of generatized DBMSs to satisfy user's
specific needs has resulted in many "custom made" Data Bases for
many users. The established trend is toward specific users and,
when needed, large "systems" to meet component needs (SAFE and
CRAFT for example).
Historical data shows a continued increase in both the number of
Data Bases and their use. In the ODP Ruffing Center the number of
GIM Data Bases is constantly increasing as shown.
Year Approximate Number of GIM Data Bases
72
2
73
3
74
7
75
11
76
15
77
19
78
24
79
.29
80
34
81
37
(See figure 1)
These GIM Data Bases are varied in scope. They cover registries,
accounting systems, resource management systems and specialized
information systems. Numbers are expected to increase even more
rapidly in the future, especially with increased registry
automation.
Use of Data Bases has also been increasing rapidly. Examples, of
this are:
CAMS is the Intelligence Community's Imagery requirements
management tasking and accounting system. Its growth pattern
is:
-3-
SECRET
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
25X1 Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Next 3 Page(s) In Document Exempt
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
-- CAMS users need access to data contained in NPIC's NDS,
however, are unable to do so from a CAMS terminal.
-- OASIS is an annotated "subset" of the NPIC NDS but is
maintained as a separate Data Base.
-- Many Data Bases contain copies of cables but there is no one
central "file" which is immediately accessible by all.
-- Data Bases containing information on various physical aspects
of the Agency exist, but not cohesively. There is no
`consolidated source or method of accessing all holdings
(phones, WPs, safes, etc.) of a particular office, component,
function, etc.
-- SAFE will redundantly contain much information that is also
held in other Data Bases.
-- Many offices maintain biographic Data Bases. Single users
often do not have access to all such Data Bases.
In general, the existing Data Bases show a strong trend toward
increased use of approximately triple in each five years (in
either number of users or actual use). Major revisions are made
when necessary, (for example, CAMS and NDS in 1984 to provide for
new/changed systems) and files and users are added as requirements
dictate.
As can be seen from the trends, Data Bases and their use continue
to grow. Therefore, the Agency must provide the procedures,
methods, equipment and personnel necessary to cope with this
expanding demand. The growth of, and required changes to, Data
Bases must be quantified for future time periods.
Along with this is needed the definition of the characteristics of
required Data'Bases. Their size, types, expected uses, and
configurations must be established. This will influence the
capabilities the Agency must provide for them.
-8-
SECRET
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Data Bases are used for specific purposes and these purposes
define their structure and the functions needed to operate them.
In order to obtain the "tools" necessary to provide Data Base
services in the future they must be defined now so that they will
be readily available when required. These could be functions such
as text searching, calculation capabilities, report writers, etc.
Also, the degree of "user friendliness" must be scoped.
With respect to this issue, the adequacy of the Agency's current
DBMS must be examined. Primarily, are GIM and RAMIS (the two
Agency DBMSs) adequate? If they are adequate but need
enhancements these should be defined. The industry now has many
DBMSs available and they could be used to satisfy some Agency
needs. Which readily available DBMSs or functions should be
implemented by the Agency must be established.
C. Issue 3 - Data Base Concepts
The current concept of a Data Base being a defined and "isolated"
collection of character data does not necessarily hold in the
future. Non-character information, foreign language Data Bases,
and Data Base interoperability are concepts that are just now
being envisioned. The requirements for these types of functions
should be defined.
The Data Base Administration function has largely been ignored by
the Agency. Because of this there are many overlapping,
redundant, and restricted Data Bases. The need for a unified
cognizant component that will "broker" data, functions, and
capabilities would alleviate much of this overall shortfall. User
services, focused requirements, and overall efficiency could be
provided by this entity.
IV. Questions
A. Growth
What are the numbers of, types of, sizes of, and characteristics
of Data Bases that are required in the next ten years?
What is the requirement, if any, for a Data Base of "external"
information (AP, UPI DJ, etc.)?
-9-
SECRET
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
What new/changed functions (services) are needed to support future
Data Bases (or current ones)? Are commercially available DBMSs
required as standard Agency "tools"?
Is it envisioned that Data Bases will be developed by vendors that
will employ industry "standard" DBMSs? If so, what provision must
the Agency make to accept them (the Data Base and DBMS) for
internal use?
To what degree is "user friendliness" a factor in Data Base needs?
Are there services/standards known to be available which the
Agency should use?
How are Data Bases to be used in the future? Are they really
"reference services", or administrative services, or operational
components of the Agency's normal activities?
C. Concepts
What are the conceptual changes expected in Data Bases? Are the
following envisioned?
-- Foreign language Data Bases
-- Non-character information
-- Efficient information transfer and/or access
How should the architecture provide for the future? Should Data
Bases be amalgamated for greater use? Fragmented? Connected?
Is there a need for an overall Data Base component to oversee
growth, functions, information needs, etc.?
What specific performance requirements apply to future Data Bases
(speed, common data, languages, etc.)?
-10-
SECRET
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Information Handling Facilities
Data Transmission, Dissemination and Distribution
I. Objectives
The objective of this paper is to focus attention on the data
communications issue from the user viewpoint. The paper is
intended to assist the working groups in defining in clear and
concise terms perceived user issues and, to the extent possible,
the user requirements for data transmission and distribution
links through the year 1989.
This paper will direct attention to the data communications
process and identify user problems that may arise as total
electronic data distribution systems become a reality. As well,
we would hope to identify perceived user requirements for
services which transcend those available from systems which are
currently deployed and/or now undergoing upgrading. The range of
user services that are projected to be available in this era will
be investigated as well as the timing for implementation.
Data Communications links are defined as the electronic
medium which enables information to be transferred from one user
to another user. To clarify and focus these considerations it is
useful to address them in terms of the following categories:
o Long Distance Data Transmission Links - Transcontinental
and/or international in scope.
o Regional Data Transmission Links - Confined to an easily
circumscribed or conveniently defined locality; e.g.,
the Washington, D.C. metropolitan area, Northern
Virginia, etc.
o Local Data Distribution Links - Generally confined to a
specific building or compound, usually not requiring
outside carrier services.
At the terminus of the data transmission links, the data
dissemination and distribution functionalities then take over
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
responsibility for timely delivery of the data to the designated
users. Within the Agency, dissemination has traditionally been
defined as the process of determining who should be made aware of
the acquired information, while distribution is defined as the
actual movement of the appropriate information to the user.
Electronic mail would provide the above services, but with the
absence of paper as the user medium.
The Agency is currently served by myriads of dissemination
and distribution "systems." Voice services aside, these systems-
service three general types of information: electronic record,
paper documents, and graphics. (For this presentation, voice
services via a secure phone at every Hqs officers desk is
presumed.) Present services to Agency field stations are further
broken down as record communications (messages plus some data,
Facsimile and interactive keyboard teleconference) and non-record
communications (Secure Voice).
A. System Architecture
A description of the Agency's current distribution system
architecture at Headquarters is presented in the diagram on the
following page. In this depiction consumers are identified at
the end of each line (without boxes) while information processing
entities, both hardware and/or software are identified within the
rectangles. Only major entities are depicted, and special user
channels and low volume linkages are not shown.
The vast diversity and the hierarchial arrangement of systems
through which the Agency's electronically recorded information is
disseminated and distributed is evidenced in this diagram. As a
current example, the Interim SAFE system is one of the most
elaborate of these systems, providing dissemination and
distribution of a wide variety of information. Interim SAFE is
served, in turn, by other systems (MAX, CDS and DATEX) which
disseminate information to it. Within Interim SAFE, users
establish profiles against which incoming information is compared
and then electrically distributed.
-2-
SECRET
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
2. MAX
The Office of Communications development and deployment of
three Message Automated Exchange (MAX) Systems in the late 60's
represented our first excursion into automated message routing of
cable traffic, this effort being mainly directed toward
supporting DDO's requirement for message services with its
overseas stations. To augment this system four fieldbased
message switches (ARS's) were later installed at OC communication
base stations overseas.
Electronically based information distribution has been served
since late 1977 by the Cable Distribution System (CDS) which
disseminates cables, based on profiles and full text analysis, in
both hard copy and electronic form. Today, about 30 percent of
the Agency's staff cable traffic is automatically distributed.
For distribution of non-Agency source traffic that is routed
through our facilities the figure is nearly 90 percent. Overall,
approximately 60 percent of the cable traffic that is processed
in the Headquarters building is electronically distributed, hence
manual dissemination is much employed in the the Cable
Secretariat's (CDS) operations.
APARS, when it becomes operational in November, will
ultimately yield a much higher statistic for the cable
reproduction and collation process, delivering the traffic ready
for branch-level distribution via the paper medium. Beyond APARS
and the four remote APARS in the DDO, however, each component is
effectively responsible for disseminating their own documents.
In addition to this, there may be further second, or even third,
order paper dissemination. This is accomplished by a myriad of
registries, Watch Officers, ICO's, etc. having no single unified
criteria used in the process. Distribution of other documents
also varies widely. Courier systems are set up and run by
various organizations, individuals and "Channels". Inevitably
there are problems with "late" mail, multiple deliveries of the
same document via different systems, and consequential
proliferation of paper. Accountability is also a continual
security problem, as is the handling of compartmented
information.
-4-
SECRET
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Graphic information is, at present, totally disseminated and
distributed manually. When ADSTAR becomes operational it will
automate much of the graphics search and retrieval process.
25X1
Most of the electronic data distribution within the Hqs
complex has for many years been satisfied by a twisted pair
(copper wire), red data grid system. This grid is presently
limited to approximately a 9,000 pair capacity, but plans are
underway to expand the grid to about 10,400 pair by the end of
CY-81, and 20,400 by the end of CY-82. The maximum expansion
capability of the Hqs red grid system is postulated to be 28 to
30 thousand pair, much of which may be realized within the next
few years. Grid systems of this nature, however, are generally
limited to a maximum reliable data rate of about 19.2 Kbps. A
similar grid has been installed and an Ames Building
grid is scheduled for installation shortly; twisted pair grids
are also slated for later installation in other Agency
outbuildings. All connections to the grid are hardwired and when
terminals are removed it is now accepted practice to pull that
connection out back to the nearest wire closet. (See the
Carrier-Rate Table for the range and type of services available
from particular carrier mediums.)
The SAFE project is bringing with it a Black-data coaxial
cable bus system which is now being installed in the Hqs
building. As shown on the carrier-rate table, this type of bus
technology has the potential to handle multi-megabit data
transfers, although the Bus Interface Unit (BIU) is currently
equipment-limited to a 19.2 Kbps rate. While the Interim SAFE
terminals are now link-serviced at from 1200 to 4800 bps, the
standard service for SAFE terminals will be 9600 bps.
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 20OUNDT CIA-RDP84B0089OR000400060016-3
Carrier-Rate Table
Usual Service Provided Rate Carrier
Teletype and very slow TTY Speeds HF Radio and unconditioned
speed data. Includes key- (75-600 bps) copper (twisted pair) lines
board teleconferencing
and off-line messages
Messages, data manipulation, Low Speed Data Low-power Satellite Radio,
data exchanges, slow speed (1.2-9.6 Kbps) Conditioned Copper (twisted
digital FAX, low resolution pair) lines
graphics, Narrow-band
Secure Voice
Data Base Transfer, High Medium Speed Data Medium-power Satellite and
Speed FAX, Slow Scan or (19.2-512 Kbps) Microwave Radio, Coaxial
amplitude compression, copper cable
Secure Teleconference,
Medium Resolution Graphics,
High Quality Secure Voice
Very large data base High Speed Data High capability satellite
transfer, High resolution (1.544-50 Mbps) and microwave Radio,
graphics (Imagery), Conditioned Coaxial Copper
"Television, High Quality Cable and Fiberoptic lines
Video Teleconference
Approved For Release 2003/08/13: CIA-RDP84B0089OR000400060016-3
25X1 Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Next 1 Page(s) In Document Exempt
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
services. In many cases, suppliers are responding to specific
user requirements without following an orchestrated plan or a
common methodology. In considering this problem, there are three
different dissemination criterion that may be considered:
-- Creator generated (information provided to A, B and C)
-- Content determined (A gets all information
containing these words, etc.)
-- User determined (search and retrieve)
Each of the above methods will be required in the future and the
capability to "examine" information once it is stored
electronically supports all three methods of dissemination very
efficiently. "Commonality" in the dissemination/distribution
process remains nonetheless a much needed capability. An
individual or component today must depend on a multitude of
methods, systems, and procedures to have his requirements
met--different types of profiles in different electronic
dissemination systems, different people and registries for
different documents, etc. Although there is, admittedly, no "one
way" to state requirements to all systems for all products, the
advantages of designing future systems to standard protocols
using compatible interface criteria from a well-defined,
orchestrated system overview would be a great stride forward.
2. Cohesive Design
Since the requirements of the sponsoring user normally
prevail in system design considerations, continuation along our
current course will likely lead to a greater diversity in our
information handling systems, each with limited perspective and
methodology. This may well result in a user-tailored and/or
user-owned system that yields to a complex diversity of overall
criteria with little thought would be given to other systems (or
related requirements) and the greater system cohesiveness that is
sought after.
3. Human Judgment
Dissemination based on human judgment is inevitably eratic
due to fatigue, environmental and personality factors. As well,
personnel changes with attendant training and security
-9-
SECRET
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
considerations stress and perturb these procedures during
transitory times. This is not to suggest that human judgement
will ever be totally eliminated from the dissemination process;
nor should it be. However, there are yet labor intensive
dissemination processes which can simply be done better by
machines, recognizing that it is oftimes difficult to consolidate
the criteria used by, or the rationale used within, many human
dissemination processes.
4. Minimize Paper
To`a large extent, planned "systems" still look to paper as
the beginning and/or ending of the process. Accordingly, the
capability to generate, transfer and deliver information
electrically is severely impaired. Principal among the
objectives to improve the information flow within the
organization over the next decade will be the effort to minimize
the amount of paper both created and distributed throughout the
organization. (Along with cable and other data traffic, this
implies that "word processors" should be tied "on-line" into a
central dissemination/ distribution(s). Compatability with and
interoperability between these systems will therefore eventually
become highly desirable, if not an outright necessity.)
5. Greater Productivity
Lastly, speed and accuracy considerations argue persuasively
for future dissemination and distribution to be accomplished
electronically wherever possible. This also supports the
increased (and ultimately exclusive) use of electronic
collection, transmission and sorting of both classified and
unclassified information so that it can be disseminated and
distributed to the individual user via electronic means. This
will promote greater system compatability and encourage wider use
of machine-assisted techniques to assist the user in doing a
better job in less time.
In the field, the spectrum of services that could soon be
made available to enhance productivity and improve support from
headquarters also represent an area that impacts signficantly on
-10-
SECRET
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
SECRET
the overall Agency ADP profile. The limited bandwidth of our
traditional data communications links with the overseas stations
has presented a severe restriction on the amount and the types of
information that could be electronically exchanged with the
field. However, over the years developments in satellite
communications and, more recently, in packet-switching technology
have revolutionized the communications industry. OC's MERCURY
program utilizes packet-switching and will make available to the
Agency a great expansion in operational and administrative
support capability. As the satellite communications carrier
services with each field station realize a commensurate upgrade
in data carrying capacity, the versatility of the packet-switched
network will permit many users dreams to become realities. Many
of these services could be available in the field by the mid-
eighties and probably all of them by the end of the decade.
2. Link Capability
On the following pages is a listing of future Field Station
service capabilities broken down into five general service
categories. The categories are keyed to the MERCURY RFP, Network
Architecture definitions and portend service--dedicated
communication links of approximately 2.4 Kbps (small stations),
4.8 Kbps (medium stations) and 9.6 Kbps (large stations). There
follows a discussion of the services that are thought to have
particular user interest, hence they are addressed individually
along with some suggestions of how these services may also apply
to information transfer within the Hqs complex. Users will be
asked to assess the value of these services with regard to their
operational requirements.
3. Field Service Listing
(Expanded Formal Message Service - many types converted to
share a common/general format using abbreviated message
format (AMF) from which accounting, logging and processing
information can be automatically derived.)
(1) Regular Command and other Wide Distribution Traffic
(policy, operations, approvals, etc. -- may include
SCI, SI and others)
-11-
SECRET
. Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
25X1 Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Next 2 Page(s) In Document Exempt
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
e. Facsimile, Slow Refresh Video Teleconferencing
(3.d.4)
The technologies of digital facsimile and teleconferencing
are, in many ways, very similar since they, too, represent vast
quantities of unformated bulk data to be transferred.
Accordingly, to be time responsive, they require a digital link
of at least 19.2 Kbps, and will then provide an image
transmission or refresh on the order of every 10 to 60 seconds.
The higher the desired resolution the lower the refresh rate, and
resolution is usually a control variable. By using slow scan or
extreme compression techniques, Black and White secure
teleconferencing can now be achieved over medium rate links as
low as 19.2 Kbps. Over time, technological advances will improve
the quality of these services. Very slow scan (every 15 seconds
or so) color services are now available over 56 Kbps carriers.
The old, AT&T continuous image teleconferencing concept has
generally been abandoned because of its impractical 3 Mbps data
rate requirement.
Digital Fax services can be provided at a relatively modest
cost making them practical for many Agency users. However, they
do require special terminal equipment at each end and a service-
dedicated channel. To provide video teleconferencing with
interactive secure voice would probably necessitate a 56 Kbps
link as well as the associated terminal equipment. As these
services become more pronounced it would be prudent to provide a
common user teleconferencing (and facsimile) centers at strategic
locations through the Headquarters complex, thus improving these
services and making them routinely available between Agency
buildings in the Washington area as well as providing a new
service to selected overseas facilities.
f. Voice Memoranda and Special Secure Voice Services
(3.d.5)
Voice memoranda are digitized voice messages that are
processed in the same store and forward manner as written
traffic. They will soon become practical as a consequence of new
electronics technology which will permit great compression of the
bandwidth requirements relevant to normal voice. The primary
usage is for daily administrative and management functions. It
permits improved professional time management through release
from the interrupt in operations enforced by the telephone's
-15-
SECRET
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
immediate-connect mode of operation. The necessity of the caller
frequently investing in a series of calls to make a single
connection is eliminated since the digitized message is stored
and later delivered when the callee is available. This part of
the process could be handled as an electronic mail type of
function.
g. Medium-Resolution Image and Graphics Capable
Terminals (3.d.6)
Data terminals which are medium resolution, graphics capable
are now commonly available in the commercial marketplace. As
soon as TEMPEST -approved versions of these devices are
developed, they will be a tool available to the Agency.
The primary applications are in data base work and word
processing. In both applications, graphics images are created or
selected, used in some context, and stored. We now have
powerful, high-resolution graphics terminals in the Agency, but
they are so expensive that their broad proliferation is
impractical. Such high resolution capability, however, is not
necessary for the bulk of the work being handled, such as graphs,
line drawings, medium resolution maps, layouts, mug shots and the
like. Low resolution requirements of this type can often be
handled satisfactorily by a 16-shade facsimile capability,
whereas medium-resolution processing requires the higher data
rates and an advanced data terminal.
V. Trends and Projections
Future systems design considerations and office-level
strategic planning exercises have prompted Agency IHS providers
to make a number of traffic analysis and growth projections.
These projections are generally derived from current and past
system usage statistics and user estimates of growth from new
requirements that have not yet impacted on the system. The
following graph depicts the growth in cable traffic volume (both
total network and agency-only levels) and the growth in data
message volume over the past six years. (The DATEX projection
does not, however, take into account the substantial impact of
the new CDS link-up; narrative traffic has been distributed via
this link only since June 1981. See Signal Flow diagram.)
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
25X1 Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
25X1
The table that follows contains the MERCURY system
projections for all different traffic types as developed in the
preceeding paragraph. These projections were derived as depicted
above using extrapolations from the approximately 11 percent
annual growth in Agency cable traffic and the 28 percent annual
growth in data traffic that has been experienced in recent years.
Moreover, these statistics reflect the order-of-magnitude
increases in general data traffic that are expected to result
once an interactive CRAFT is serviced by a full-capability (in
phases) MERCURY and an expanded SKYLINK system. They also
reflect the reduction in traditional cable traffic volumes that
is expected, since much of the information transfer that is now
made in this format will be more efficiently conveyed via MERCURY
as preformatted data messages. Considering all categories, the
overall growth in data transmission requirements once interactive
sytems are on line is, indeed, going to be impressive and
challenging.
The issue statements below are to provide a general frame of
reference from which the users can examine the specific questions
in the following section. Although the questions are arranged in
order (relative to the issues), an appropriate response may, in
some cases, relate to both issues.
Issue A - DATA DISSEMINATION AND DISTRIBUTION AT HEADQUARTERS
ASSURE THE ADEQUACY OF FUTURE DATA DISSEMINATION AND
DISTRIBUTION SYSTEMS TO SATISFY THE PERCEIVED USER NEEDS OF THE
1980's. EXAMINE AND EVALUATE CURRENT AND PLANNED DISSEMINATION
AND DISTRIBUTION SYSTEMS AND, AS REQUIRED, RECOMMEND APPROPRIATE
ADJUSTMENTS TO FACILITATE THOSE SYSTEM IMPROVEMENTS THAT WILL
SATISFY USER TRENDS AND TAKE ADVANTAGE OF EXPECTED TECHNOLOGICAL
DEVELOPMENTS.
Issue B - COMMUNICATIONS SERVICES WITH THE FIELD
FROM A COMPREHENSIVE EXAMINATION OF THE INFORMATION
PRESENTED IN THE LAST PART OF SECTION IV.C, DETERMINE WHAT RANGE
OF IH SERVICES ARE MOST REQUIRED AND ARE OF GREATEST UTILITY FOR
FIELD DEPLOYMENT BY THE YEARS 1985 THRU 1989. EVALUATE SERVICE,
COST. TIMING. E-OFF FACTORS AND CONSIDER THE IMPACT OF
QUIREMENTS WHERE APPLICABLE.
-18-
SECRET
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
25X1 Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
A. What are the problems that users see with the
dissemination/ distribution system that is evolving here
at Headquarters? What is the preferred criteria for
dissemination at the various levels? How does "need to
know" (and other security considerations) affect this?
b. What problems do users have (if any) with the development
of a standard message format for Agency message traffic?
c.= Should the information dissemination and distribution
system be expanded to accommodate unclassified wire, news
and data sources?
d. What will be the future applications for local networks?
Why will they be required?
e. How should we handle the mechanics of terminal access in
the transition period from relatively few terminals to a
time where terminals are readily available to all
employees who can use them? What are the relative merits
of personnel dedicated terminals over personnel-movable
shared terminals? Over fixed -location, shared
terminals? How much appeal would a rapid-access, plug-in
dataport have to users? How much validity does the old
"terminal room" concept have?
f. What will be the time-phased demand for committing all
cable and data message traffic to electronic distribution
medium? To what extent will paper channels still be
required? At what intermediate level would this
information be expected to enter/exit the dissemination/
distribution system?
g. What is the time-phased demand for interactive links with
field stations? Between field stations?
h. What will be the time-phased demand for electronic
distribution of the many regulatory issuances, notices,
bulletins, instructions, etc. that are now delivered via
pouch/mail channels? Will hard copy always be a
requirement for certain classes of documentation?
-20-
SECRET
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3
i. What will be the time-phased demand for facsimile,
teleconferencing and graphics services requiring medium-
speed (19.2 - 256 Kbps) data rate link capability in the
field? What growth in requirements of this type is
expected locally (between buildings in the Headquarters
area)?
What will be the time-phased demand within the
headquarters community for work station imagery, secure
television or other services requiring high-speed (1.544
Mbps and more) data links.
k. In terms of user requirements and desirability, which of
the field station services listed in section IV.C are
most urgently needed? Are there any services presented
which are not justifiable?
-21-
SECRET
Approved For Release 2003/08/13 : CIA-RDP84B0089OR000400060016-3