SAFE PROGRAM STATUS REPORT

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP84B00148R000200440033-1
• Release Decision: RIPPUB
• Original Classification: S
• Document Page Count: 45
• Document Creation Date: December 20, 2016
• Document Release Date: June 12, 2007
• Sequence Number: 33
• Publication Date: June 22, 1982
• Content Type: REPORT

Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
SAFE PROGRAM STATUS REPORT
June 22 - July 22, 1982
TABLE OF CONTENTS
Page
Preface. . . . . . . . . . . . . . . . . . . . . . 1
II. Background . . . . . . . . . . . . . . . . . . . . . 2
III. Evaluation Phase Tasks . . . . . . . . . . . . . 5
A. DBMS Study. . . . . . . . . . . . . 5
B. Other Software Package Study... . . . . . . . 12
C. User Language . . . . . . . . . . . . . 21
D. Requirements Closure. . . . . . . . . . . .-. . . 23
E. Conversion Analysis . . . . . . . . . . . . . . 24
F. Systems Engineering Analysis. . . . . . . . . . . 25
.IV. Schedule . . . . . . . . . . . . . . . . . . . . . . . 26
V. Costs. . ... . . . . . . . . . . . . . . . . . ... . . 30
TABLES:
I. Model 204 (DBMS) Deficiency Estimates (follows page 11)
II. Electronic Mail Software Packages (follows page 17)
III. Task Summary Schedule (follows page 27)
IV. Cost/Funding Profile Summary (follows page 31)
V. Cost/Funding Profile (Software Integration Option)
(follows page 31)
VI. DIA/CIA 1984 Program Submission (follows page 31)
SECRET,
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
 crrh
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
SAFE PROGRAM STATUS REPORT
June 22 - July 22, 1982
PREFACE
The purpose of this document is to provide a status
report on those SAFE activities since the 22 June 1982
meeting of the SAFE Steering committee. At that meeting,
the Consolidated SAFE Project Office (CSPO) was asked for a
30 day status report of its activities to include, to the
extent possible, insight into the software cost
specifically with respect to the augmentation of any
available software packages.
This report is organized into five sections including
this preface. Section II provides background material that
speaks to the precepts upon which the proposed redirection
has been built. Section III provides information on the
status of the -seven evaluation phase tasks and Sections IV
and V discuss the projected schedules and costs
respectively.
1 SECRET
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
II. BACKGROUND
The approach to the development of SAFE adopted by the
.Consolidated SAFE Project Office (CSPO) recognizes the
strong relationship that SAFE has to commercial data
processing developments in the.areas of office automation,
electronic mail, and data base management systems. The
approach seeks to determine how the marketplace may best
.support the SAFE.. requirements. while achieving an integrated
environment that is easy for the intelligence analyst to
use. This is an interim report of progress, made since 22'
June to evaluate this approach.
Fundamental to the development of SAFE is 1) maximum
use of available software and 2) incremental delivery of
capabilities to the users. In order to make optimum use of
applicable commercial and government software, an
IBM-compatible hardware environment is to be pro'rided.
Communications support to the terminals will be by non-
encrypted twisted wire pair in order to minimize cost and
risk. Twisted wire point-to-point communications already
exists at the CIA and is being installed in the new DIA
building. Tasks performed during the evaluation phase will
result in the selection of software packages to be used as
part of SAFE. Each of the software selection tasks
includes the determination of what augmentations (and their
costs) are required to these packages in order to provide
SECRET
Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
the. requested capabilities. Additionally, the studies
determine how to put the software packages together to
provide the combined set of services. Further, all of the
software must function with the computer and communications
hardware. The design for these integrated capabiliti.es
will be presented at a system design review in February
1983.
The costs and schedules to develop these augmentations
and the risk of being able to achieve them will be
assessed. High risk or high cost augmentations will be
weighed against their value to determine if they-should be
developed. This process also applies to the issue of
achieving the integrated user-interface. Where it is
-deemed that the risks or 'costs are excessive relative to
the value received, augmentations to achieve the user
interface. integration will not be attempted. . The
assessment of the value of augmentation will be made at
appropriate intervals, primarily at design reviews
throughout the project.
We expect the marketplace to be active in solving many
SAFE-like problems through improved performance hardware or
through software with greater capability. An incremental
approach to the development of SAFE allows continuous
monitoring of the marketplace.
SECRET
3.
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
Once the software package selections, augmentations,
costs and schedules have been determined, a project plan
for SAFE can be'established. This will be available in
November 1982. Additionally, a management plan will be
provided in August 1982. The management plan will address
how this type of multivendor software acquisition,
development and integration can be performed as well as how
the government works with the contractors. The CSPO has
organized itself into three segments: system development,.
quality assurance, and operational systems.
The CSPO is currently developing the detailed plan for
providing an early capability. The plan will be available
by 16 August.. This plan will provide for opening.the SAFE
computer center at CIA in March of 1983. It will contain
three facilities: an expanded Pilot Mail Operation for
CIA, a separate capability for the DIA (using DIA message
traffic), and an unclassified development facility (for
integrated capability development).
Since the Steering Committee meeting of 22 June the
CSPO has concentrated on specific tasks critical to the
evaluation of the proposed development approach. The
objectives of each of the evaluation areas and the status
of reviews are summarized below. This evaluation process
will continue through September.
SECRET
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
III. EVALUATION PHASE TASKS
A. DBMS Study
The objective of the DBMS selection subtask is to
select an IBM compatible Data Base Management System
(DBMS) and text handling package that best meets the
needs of the SAFE project. The; selection is based on
adherence to SAFE requirements as defined in the
selection criteria. The DBMS is to be selected from
the broad spectrum of commercially available products
on the basis of their. inherent capabilities and their
compliance with SAFE functional, maintenance, resource
usage and environmental requirements.
If existing software does not . cover the
requirements, the study will identify enhancements to
existing software and/or the need for additional
software required to satisfy the SAFE Data Base
Management System requirements.
The technical. approach for evaluating and
selecting DBMS and text packages insures that the
systems currently available are considered and makes
certain that no promising candidates are omitted.
Furthermore, the approach insures that the systems of
S , SECk~ i~
Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1 ----
 Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
interest are subjected to a sufficiently exhaustive
technical analysis to validate the recommendations and
conclusions.
The technical approach to DBMS performance
assessment includes effort to design a prototype data
base that will exercise the important file structures
and relationships for both the CIA and DIA customer.
There will be two data bases defined, one for textual
data and one for structured data.
An extensive literature search has been conducted
which included. the DATA PRO and AUERBACH reports, a
DATAMATION listing and various internal TRW
compilations. The list was screened to select those
DBMS packages that are compatible with an IBM
environment. This resulted in the identification of
19 structured DBMS's. Separately identified were a
list of 14 possible text DBMS's.
SECRET
6
Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
The candidate DBMS's are as follows:
Structured DBMS
ADABAS
BASIS
DATACOM/DB
DL/1
GIM II
GIM III
GIS/2
IDMS
IMS/VS
INQUIRE
MODEL 204
OASIS
ORACLE
RAMIS II
SEED
SIBAS
SYSTEM 2000
TOTAL
SQL/DS
TEXT DBMS
BASIS
CONTEXT C-705
COSMIC
DIALOG
DOCU/MASTER
IMDOC
INQUIRE
LEXIS/NEXIS
ORBIT III
STAIRS/VS
SUPER FAST STORAGE
AND RETRIEVAL SYSTEM-
TEXT 204
These DBMS's were- analyzed to identify the
will undergo detailed
technical evaluation. This was accomplished by
SECRET
17
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
applying a set of mandatory requirements to the
previously identified IBM compatible packages and then
performing a preliminary technical evaluation where
necessary to reduce the list to the following:
Structured DBMS
ADABAS
IDMS
INQUIRE
MODEL 204
SYSTEM 2000
Text DBMS
BASIS
DOCUMASTER
INQUIRE
STAIRS/VS
.TEXT 204
familiarization with. the DBMS's selected for
evaluation has been started for all packages
and has been completed for:
ADABAS
IDMS
INQUIRE (structured and text)
MODEL 204 (structure and text)
SYSTEM 2000
The remaining text systems are still under study.
8 :SECP-,E i
Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
The detailed technical evaluation has progressed
significantly against the following packages:,
ADABAS
INQUIRE (structured and text)
MODEL 204 (structured and text)
SYSTEM 2000
.Technical meetings were held with the vendors of
each of the above packages to gain a complete and
detailed understanding of the internal data structures
and data access methods employed and to resolve other
technical issues that were open as a result of
reviewing the technical documentation. Initial
performance projections were made based on the
information gained.
- r-..
Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
 rb t--
Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
Cost data has not been captured yet for most
systems but representative data is shown for Model
204:
Acquisition Cost:
Basic Package 65K
User Language 40
Host Language Interface 30
Utilities 20
One T.P. Interface. 15
Data Dictionary 20
Math.Pack 5
Total license fee 195K for the first
copy
There is a 20% GSA discount on the license fee
only for government installations. Additional
discounts may be available for multiple copies. A
minimum of two copies per Agency will be required for
an estimated acquisition cost of approximately $650K.
.At this point in the evaluation process, it is
premature to give definitive answers concerning cost
and the degree to which SAFE requirements can be
met. Given the fact that the DBMS decision will be
one the analysts will have to live with for the next
fifteen to twenty years, the goal is to select the
best available package based on an overall evaluation
SECR n-
Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
SECRET
which includes what needs to be done to overcome the
particular shortfalls of the package. The available
list of candidates has been narrowed, however, and it
is possible to describe a representative case, which
will be covered in much more detail as the study
progresses. Model 204 is one of the DBMS finalists and
significant research has already been done on this
product. Although M204 may not be the chosen package,
it is representative of .the class of available DBMS's
that would come closest to meeting SAFE
requirements. As previously stated, no single DBMS
will fully meet our needs. The, following is a
discussion of these areas where M204 is deficient and
a possible approach to overcoming the deficiency. if
a DBMS other than M204 is chosen, the actual
deficiencies may be different but nevertheless similar
in that a certain important subset of the requirements
will not be met.
Capabilities not supported by Model 204 and
preliminary estimates of associated cost and risk are
shown in Table I, which is an example of the kind'of
output that will be included in the final report on
all software package evaluations. If we decide to
implement the changes needed to meet all of the
identified DBMS requirements, the cost is estimated to
be from $2M to $3M.' It should be noted the original
SECT r
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
Shortfall Cost ($) Risk
Geographic Search - Search on coordinates
for inclusion within specified geometric.
figure (i.e., polygon). 141 K low
Maintenance via Output Forms - update of
record content using output form display
on screen. 71 K medium
Transaction Files - ability to collect
?update transactions in a file and is
cause deferred execution of file
content. 71 K low
Archiving - Deletion of aged documents
from online files, storage on tape,
ability to subsequently search and
retrieve from tape.
71 K ~ low
Distributed Data Base - Ability to search
and update multiple files that are
distributed across several processors
in a loosely coupled network. 282 K high
Integrity Checking - to verify correct-
ness of internal pointers throughout the
data base. - 71 K medium
Sync Point Recovery - To be able to
recover the data base to the last
completed transaction from the last known
consistent state. 141 K medium
Multifile Transaction Recovery - To be
able to recover the data base to a
consistent state when transactions are
involved that apply to more than
one file. 141 K high
Field Level Security - To allow field
level access consistent with
compartmentation restrictions. 71 K low
SECRET
Table I (page 1 of 3)
Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
-&_%vI%I..^
Shortfall Cost ($) Risk
Data Dictionary - the following are
shortfalls that can be corrected by
the implementation of a separate
data dictionary. 352 K medium
Unsupported Data Structures:
o Subfields - subdivisions of an
already defined field (i.e., MO
in DATE: MODYYR)
o Repeating groups - logical collection
of fields within a record that may
consist of several instances.
o Index term phrases - collection of
fields that have a defined
relationship to one another.
o Non-contiguous zones - text fields
that consist of several physically
separated parts.
o Overlapping zones - text fields with
a non-empty intersection..
Unsupported Data Dictionary Functions:
o Field level validation - verify
correctness of data entered in a field.
Triggers - when updating a field, the
ability to cause related updates to
occur elsewhere in the data base.
o File level status - ability to collect
and store statistics on files 'in the
data base.
SECRET
Table I (page 2 of 3)
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
MODEL 204 DEFICIENCY ESTIMATES
Shortfall Cost ($) Risk
Linked file hit list - The hit list
resulting from a linked file search,
(e.g., "Find full professors and
their straight 'A' students") should
(but does not) preserve inter-record
relationships. 225 K high
Linked file search response time - M204
is much slower than required. 254 K high
Multi-file search - Capability of
searching several dissimilar files and
concatenating the results. 56 K medium
Dynamic browse - Records added to a file
after a user starts displaying it. Should
be (are are not) displayed. 113 K medium
Skip a record - Capability to skip a
specified number of records when browsing
(displaying) a file. 56,K medium
Fast range search - E.g., "Find employees
whose salary is between 1000 and 2000."
Response time requirement for range search
cannot be met. Specific problems are: 564 K high
o Character range search is very slow
Embedded and trailing don't cares (e.g.,
"Find all names starting with J") are very
slow.
o Numeric range search is.slow.
o Geographic search (e.g., "find all
cities within 100 miles of
Washington") is slow..
Table i (page 3 of 3)
Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
 CGr"n r- r
Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
development cost of Model 204 is many times these cost
figures. It is an advantage of the software
integration development approach that only a small
fraction of the original development cost is paid by
the SAFE program as part of the package acquisition
B. Other Software Package Study
The objective of this task is to identify which
packages will be used in addition- to the packages
selected by the' DBMS Study ,task to provide an
integrated SAFE system... We will evaluate both
government and commercial packages to determine which
can best satisfy the functional requirements of
SAFE. If the existing software packages do not cover
the requirements the study will identify enhancements
and/or the need for additional software required to
satisfy the SAFE requirements. These three functional
package. studies cover the requirements for Mail
Dissemination, Non-Mail Applications, and Operating
Support Software. In a parallel Software Architecture
study we will coordinate the activities- of- the DBMS
study and the three functional package studies and
describe how the selected packages will be combined
into an integrated SAFE system. The Software
Architecture effort will document the design in the
:SECRET
17
Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
software portion of the System Design Document. The
document will include a traceability of all SAFE
requirements to either existing software packages,
augmentations to those existing packages, operating
system software or custom software. In addition, the
document will provide a description of package
interfaces and sequencing, and a technical discussion
of both local and.system wide issues.
The Mail Dissemination study focuses on packages
that can provide SAFE mail dissemination. This basic
mail capability will include the following functions:.
Message analysis
The ability to receive electrical messages and
interpret and decompose them into zones
according to predefined rules.
Mail dissemination
The ability to match incoming messages against
user interest profiles and deliver mail based
on those profiles.
Mail viewing
The ability to format mail for display or
hardcopy and to dispose of mail in a variety
of ways.
ECRU"
13
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
Indexing
The ability to associate user-defined keywords
with a given message and use those keywords
for subsequent retrieval of the desired
message.
The Non-Mail Application study will focus on
packages that provide functional support for a SAFE
user. This. study will include evaluation of the
following functions:
- Text editing
- Word processing
- Special file maintenance
- User interface presentation
- HELP/CAI (Computer Aided Instruction)
- Terminal interface
- Command language parsing/translation.
The Operating Support Software Study will focus
on packages that provide functional support for the
SAFE operators and for the package integration. This
study will identify which package, including the host
Operating System, will be used to provide the
following capabilities:
- Initialization/Termination
- Operator interface
- Intra--and inter-machine communications
SECRET
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
 %F1 WI-Y_
Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
- Failover
- Error collection and reporting
- Health monitoring
- Security processing
- Tape management
- Management Information System (MIS)
- Print job control
- Hardware configuration
- Terminal interface
- ODP link
The Software Architecture study will complete the
system design. It will investigate the candidate
operating system to resolve potentially critical
design issues and to validate the allocation of
functions to .-the operating system. The complete
system design will be compiled from the DBMS, mail,
non-mail applications and operating support software
inputs. Potential functional overlaps, interface
discrepancies, and requirements gaps will be
identified and resolved. In conjunction with the
system engineering studies on design validation
(below), we will trace the requirement. specifications
to specific software packages. The integrated system
approaches to critical issues such as data integrity,
recovery, security and user language processing will
be documented in the System Design Document along with
.SECRET
15
Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
operating system issues such as network communication,
file sharing and terminal interface.
SECRET
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
0 L..1*-.I N L. a
1. Mail Dissemination Study
We have identified 14 mail packages, twelve vendor-
provided and two government provided. A list of all 14.is
contained in Table II. We have established contact between
TRW/CSPO and all vendors and' have requested written
materials, and we have had technical presentations
conducted with SISC (MAILBOX), IBM (PROFS), CCA. (COMET204) ,
and Government (AIM, PMO).
Meanwhile, we have conducted the first pass evaluation
of.all 14 packages and selected seven packages (five vendor
and two government) for submission to the second pass
evaluation. They are:
COMET204 - TOSS
MAILBOX (STSC) - AIM
- OMNICON - PMO
- PROFS
SECRET
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
SECRET
TABLE II. ELECTRONIC MAIL SOFTWARE PACKAGES
AIM
Agency (in-house development)
Ron H.
MAILBOX
I.P. Sharp Associates
Los Angeles, California
(213),277-3878
Peter Forman
COMET204
Computer Corporation of
America
Arlington, Virginia
(703) 522-1717
Chuck Tuebner
CYBERGRAM
Cybertek Computer Products, Inc.
Culver City, California
(213) 649-2450
Mary Cope
EMAIL
ADR
Princeton, New Jersey
(201) 874-9100
Jackie Cullins
INFOMAIL.
BBN Information Management
Cambridge, Massachusetts
(617) 497-2929
Bob Harvey
INFOPLEX
Compuserve, Inc.
Columbus, Ohio
(614) 457-8600
Lynn Galli
MAILBOX
Dialcom, Inc.
Siver Spring, Maryland
(301) 588-1572
Ken Norris
MAILBOX
STSC, Inc.
Bethesda, Maryland
(301) 984-5100
Richard Paulson
OMNICON
On-Line Software International
kiver Edge, New Jersey
(201) 488-770
Mariel Cruise
PMO
Agency (In-house Development)
Gene C.
PROFS
IBM
Los Angeles, California
(213) 642-1490
Bruce Barton
QUICK-COMM
GEISCO
Arlington, Virginia
(703) 276-4053
Don Henn
TOSS
National Business Systems
Hartford, Connecticutt
(?203) 677-8396
Jy Chang
SF(-Der
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
The following areas of functional requirements were met by
the seven selected packages:
- Message Analysis
None
Mail Dissemination
PMO
- Mail Viewing
COMET204 TOSS
MAILBOX ADM
OMNICON PMO
PROFS
Indexing
MAILBOX TOSS
OMNICON AMD
PROFS PMO
The following preliminary cost data shows the range of
prices.for this kind of software: .
COMET 204(CCA) - $57K/computer if purchased alone
$35K/computer if purchased with
DBMS M-204
- MAILBOX (SISC) - $40K/computer
- PROFS (IBM) - No purchase cost; rental $370/mo
for each computer
- TOSS (NBS) - $7,200/computer
SECRET
Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
2. Non-Mail Application Study
The survey of available packages has produced an initial
set of eleven which cover the following areas of functional
and derived requirements:
a. Text editing/formatting
ATMS: An IBM product which supports text
entry, editing, formatting, , storage and
retrieval of documents. This package
supports an interface to STAIRS (text
search package) and the IBM DCF formatter.
- DCF: (Document Composition Facility): an
IBM product which provides text formatting.
DLF: (Document Library Facility): an IBM
product which provides document storage and
retrieval by name.
WYLBUR: A Rand Corporation product which
provides text editing with extensive macro
features.
SECRET
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
b. Command language/user interface/maintenance
QBE:' (Query by Example): an IBM product
which provides an interface to support
searching, sorting, creation of files,
maintenance and browse.
IXF: (Interactive Extension Facility): an
IBM product which provides a general purpose
menu-driven interface to application
software.
C. HELP/CAI
Interactive Instructional Authoring System:.
an IBM product for generating CAI-type
instructional material.
Interactive Instructional Presentation
System: accompanies the above to provide a
system for presenting instructional-material.
d. General Packages
SyncSort: produced by SyncSort Inc. as a
general sort/merge package.
DMS (Development Management System): an IBM.
product which provides support for developing
applications to run under CICS.
IMS/ADF: (Application Development Facility):
an IBM product which provides support for
developing data base applications.
SECRET
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
 C Ct rnrr
Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
C. User Language
A User Language Specification (ULS) was produced and
'delivered on 10 May 1982. It contained the majority of the
user interface and commands for the CIA user. The emphasis
of these activities is to complete this work for the DIA
user. The set of user interface issues to be addressed for
DIA and the approaches to be taken for each, are being
generated and coordinated with CSPO before detailed updates
to the ULS are performed. These additions are to be
derived from the DIA unique functions identified in the
system requirements specification.
addition to the- ULS baselining, 'a set of user
interface scenarios will be prepared to demonstrate and
validate the language. These scenario definitions will be*
a joint effort of TRW and CSPO, and expand the set already
prepared by TRW.
Preparations for the baselining of the ULS are well
underway. A list of the DIA capabilities that need to be
addressed for the ULS has been generated and implementation
approaches are being considered. Two of these approaches,
for linked file search and geographic search, have been
prepared and will be coordinated with CSPO. An analysis
has been performed to determine what issues must be
resolved before the user interface can be implemented.
CRT
Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
 Crrl)rT
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
These are subject to coordination and prioritization before
updates to the ULS are generated. A preliminary analysis
of the software studies has been performed to determine a
list of areas which are highly subject to change based on
the pending selection of vendor software packages.
Software packages being evaluated in the DBMS, Mail, and
Non-Mail Studies will be analyzed to determine their
ability to provide the SAFE user interface. If existing
packages do not cover the requirements, enhancements or
translations will be identified. This task will support
the evaluation of appropriate software packages, support
the identification of enhancements, and support
identification of cost schedule and- risk assessment
associated with each enhancement.
SECRET
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
D. Requirements Closure
The objective of the Requirements Closure Task is to
update and baseline the system level requirements. The
technical approach includes coordinating all, currently
identified requirements issues pertaining to these
specifications with the CSPO and update the specifications
to reflect the corresponding changes. Incomplete
requirements in the areas of the DIA communications systems
interfaces will also be specified. The requirements will
then be allocated for implementation in a specific
delivery. . As continued software package evaluation and
trade-off analyses. are performed and requirements changes
.are directed by the CSPO, the specifications will be-
updated to reflect these changes.
C
SECRET
Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
 .- r-. o- r
Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
E. Conversion Analysis Support
The objective of Conversion Analysis Support is to
specify the requirements for data base and program/product
conversion (conversion of the DIA online system data base
and functional conversion of the DIAOLS programs/products)
and to define the corresponding design approach which
implements these. requirements. Existing programs/products
and data. will be analyzed to verify that they are
appropriately described in the Conversion Requirements
Specification (CRS). A conversion plan will then be
developed to identify all necessary conversion software.
The relationship between this task and the software
evaluation and architecture subtasks is. significant in that
the selection of. software packages can impact the
conversion approach and the amount of conversion that must
be performed for the DIA system. Also, the relationship
between this task and the user language task is crucial
because a large amount of the conversion activities will of
necessity be performed by means. of command procedures.
SECRET
Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
~tLKt I
F. Systems Engineering Analysis
This effort will identify critical design areas which
need to be addressed by the overall design and establish
the system level architecture based on the software
evaluation studies and resolution of the critical design
issues. It will define the hardware architecture and
establish design concepts for all external and device
interfaces. Then the task will validate the overall design
to ensure that the requirements are covered, the design is
end-to-end complete, and that potential performance
problems are identified and resolved. it will also
coordinate the documentation of the design in the System
Design Document (SDD) and coordinate the preparation and
conduct of the System Design Review. A preliminary system
test plan will also be developed to identify. the
integration and test approach which will be used for the
Integrated. Capabilities.
We have formulated a preliminary list of critical design'
areas. TRW personnel attended discussions on AIM with ODP
and CSPO to learn more of current system architecture. The
current architecture for systems that will form a part of
.the SAFE early capability system as well as other similar
packages currently operational within the ODP computing
center is being analyzed. Information received to.date on
the architectures of AIM, PMO, TADS and CAMS II revealed
SECRET
Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
 T:T
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
several problems which need resolution in order that the
SAFE system can serve a larger user community.
preliminary paper has been prepared which focuses on the
architectural questions that need further investigation as
potential enhancements, alternatives, or design changes to
serve the larger user population are considered.
It is clear that there is a wide range of possible
architectures to pursue. We?plan to prioritize some major
architecture issues in order to reduce the range of
alternatives. The software evaluation criteria will be
reviewed with respect to potential architectural
environments in which the software may reside. A list of
critical system design issues will be published to
coordinate the system design activity.
.IV. SCHEDULE
The development schedule has been divided into several
phases. The first phase is for study prior to the generation
of a system design. It is called the evaluation phase. During
this period, July-Sept 1982, software packages for use in SAFE
shall be identified as well as the augmentations to them
required to achieve the needed functions. The phasing of the
functional deliveries as integrated capabilities will also be
determined.
SECRET
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
Table III illustrates the proposed scheduling of The
Evaluation Phase tasks. This is followed by a system design
phase, Sept 1982-Jan 1983. During this period the system
design will be generated based on the results of the study
during the evaluation phase. A project plan will be developed
by November 1982. It will contain.detailed costs and schedules
for the project. A formal presentation of the design, costs,
and schedules will occur in Feb 1983 at the system design
review. At present four deliveries of integrated capabilities
are proposed. They are discussed below.
Occurring simultaneously with the evaluation phase is
planning for an early capability. This capability is off-the-
shelf with very little development. It primarily provides
expanded user access to the currently installed Pilot Mail
Operation (PMO) for the CIA. The number of users will be
increased from 45 to 300; 150 users will be able to access the
system simultaneously. PMO, developed by Chase, Rosen, and
Wallace, Inc. (CRW), draws upon capabilities of Interim SAFE.
It provides a partial version of the mail operation intended
for SAFE. PMO, in conjunction with ODP's Automated Information
Management (AIM) software package will provide both mail
support functions and electronic mail (message routing). The
AIM software is supported in the %r?1/CMS operating system. This
operating system provides text editing, text composition, and
file handling services. PMO runs under the MVS operating
system. All this software operates on IBM-compatible
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
TASK SUMMA., Y SCHEDULE
TASK 1
O6MSSTUOY
TASK 7
tl OTHER SOFTWARE -ALKACE EVALUATION STUDIES
SUBTASK 7.1 EVALUATION CRITF MA
SUBTASK 7 7 NON MAIL APPLICATIONS STUDY
SUBTASK 71 CS:/HTWARF EVALUATION STUDY
SUBTASK 74 SOFTWARE ARCHITECTURE
TASK4
USER LANGUAGE
SUBTASK 4.1 ULS BASELINE
SUBTASK t 7 PnOTOTYPING
SUBTASK, 7 USFR LANGUAGE INTEGRATION
STUDY
TASK S
RTOUIRFMENTSCLOSUnE SUPVORT
SUBTASK S I SYSTEM RFOUIREMENTS
SPECIFICATION
SUBTASK 6.7 INTFRFACF SPECIFICATION
UI ROMTS.
REVIEW
BOOKS t S, 7 3
BASELINE
INPUT FROM
CSPO
i I 1.1 I I i I I l 1 1 i ..
i I II_ IN
FROM
7.1. 7.3.
7.1. 7.6
TO 7.7,
7.4.76
A 7.7
EnOM
7.6
FROM
7.6
' l l 1 1 1 1 1 1 1 1 I I
TO 7.1, 7.7.
1.4. 1.8
SO 7.7
Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
7.7. 7.7.71. 7.5.
7.8. 7.7 A 7.9
 r)
TASK 5
CONVERSION ANALYSIS REPORT
SUETASK5.1 CONVERSION REQUIREMENTS
SFE CIFICA T ION
SUBTASK 5? SUPPORT ANALYSIS OF EXISTING
PROGRAM
SUBTASK I3 .OfSIGN
TASK 7
SYSTEMS ENGINEERING ANALYSIS
SUBTASK7.1 SYSTEM ARCHITECTURE
SUBTASK 7 7 HARDWARE CONFIGURATION
SUBTASK 13 If- DTSIGN
SUBTASK 7.4 DESIGN VALIDATION
SUBTASK 7.5 PERFORMANCE ANALYSIS
SUBTASK 7.5 IAT PLANNING
SUBTASK 7.7 FUNCTIONAL OVERVIEW
SUBTASK 7.8 SOD PREPARATION
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
IS
TASK SUMMAmY SCHEDULE (Continued)
Oil
III 11 1 1 1 1 11
LI IIII I
~U 1IilL 11
III I~III II
S L
I
11
Q'. id Id o 0
L Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
F ROM
5.1 b FRO
/.
U
7
5.? ?1
FROM
BUUKS ib7 FROM
74
INPUT
FROM
F-0-1 T~ FROM
1.5.7
FROM
CSPO
TO
S.7
FROM
?A
C
C
I (
16
S
IB
18
is
1B
IB
1
is
1
IS
TO
SDR
n
O
 Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
hardware. CIA message traffic to this system will be provided
from the CIA Communications Center. All CIA terminals will
have the capability of accessing the Ruffing Computer Center.
COMTEN communications switching hardware will be used
.achieve this capability. A duplicate of this service will be
provided to the DIA. Until the new DIA building is ready for
equipment installation, the DIA capability will be housed in
the CIA SAVE computer center, but separate from the CIA
machines. Access to the DIA message traffic will be provided
by providing telecommunication lines from the -Pentagon.
Terminal access. will be from Arlington Hall. Station and the
Pentagon. A third facility to be housed in the CIA Computer
Center, during this phase, is an unclassified (and therefore
separate) development system for the integrated capability.
The project plan for the early capability will be'completed
by 16 August 1982. A detailed design review is scheduled for
the week of 27 September 1982. The early capability will be
operational in March 1983.
An enhanced capability is scheduled to be operational by
October 1984. As described in TAB A this has been viewed as a
follow-on to the early capability in which the PMO and AIM
software is augmented. Specific enhancements include: the on-
line text search software (COLTS), from Interim SAFE, is moved
from the MVS operating system to the V1.1/CMS operating system,
91:rD PT
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
 a \
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
an .enhanced word processing capability is provided, a prototype
of the SAFE User Language interface is provided, PMO is
integrated with AIM under the VM operating system, an improved
me.ssag.e dissemination algorithm is provided, and the number of
users is expanded to 500 in both the CIA and DIA systems.
Integrated Capability One is scheduled for March 1985. In
this delivery an inverted index search for structured data is
provided. Enhanced mail analysis, algorithms are available.
Conversion of the CIA's central index (RECON) is provided as
well as conversion of the DIA's structured file environment
(specifically the AIF, DIOBS, and Crisis Management file
systems). An integrated user interface is provided.
Integrated Capability Two is scheduled for December 1985.
This will include combined text searching and structured file
searching for mail as well as inverted text file, searching for
retrospective intelligence analysis activities. The number of
users. supported is expanded to 1000 for each Agency.
Conversion of the installation files and order of battle files
will be completed.
In Integrated Capability Three, scheduled _for December
1986, additional functions and modifications to the previous
capabilities are provided. Conversion of the DIA files
continues with incorporation of the CRD, HUMINT, and SIGINT
file systems. User experience with the previous deliveries
SECRET
29
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
 C rf` 1) r-T
Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
will generate new requirements which shall be reflected in this'
and subsequent deliveries.
With the delivery of Integrated Capability Four, in
September 1987, it is expected that all SAFE capability
previously contracted for will have been provided. SAFE
development should subsequently reflect normal system.
operations and maintenance activity. The level of this
activity will be directly related to changes in the missions
and functions of the user organizations.
. V. COSTS
The cost profile (Table IV) for SAFE is based on the best
estimates available from the analysis of the integrated
software approach. It was developed by discussions with TRW,
CIA/ODP, and by drawing on the CSPO (CIA/DIA) experience gained
over the past three years of SAFE development.
The FY 83 funding level is within the President's proposed
budget now before Congress. The remaining profile is under
funded by $28 million when compared to the President's FY83
budget but is within the program submissions of both agencies
for FY84-FY88.
SECRET
Approved For Release 2007/06/14: CIA-RDP84B00148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
Table V is a breakout of cost for the software integration'
approach to SAFE implementation. Emphasis is on reducing costs
associated with major software development and shifting those
costs into hardware procurement. Under the previous approach
to SAFE implementation (major software development) software
costs were more then double the $30 million and hardware costs
were less than half the $75 million.-
Table VI provides funding profiles by Agency as reflected
in the FY84 program submissions.
SECRET
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
 crrt>T
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
(Table IV)
0
COST/FUNDING PROFILE SUMMARY
(IN MILLIONS OF 1982 DOLLARS)
4th Qtr
FY82
FY83
FY84
FY85
FY86
FY87
TOTAL
IBM OPTION (BUY/
PURCHASE PLAN)
1.5
32.0
29.0
31.0
24.5
17.0
135*
DIA/CIA
PRESIDENT'S
FY83-87 BUDGET
6.0'.'..
'32.1
16.7
19.8
16.5
16.2
107.3
DI
A/CIA
FY84 PROGRAM
6.0
32.1
29.1
.30.9
25.4
18.2
141.7
*SOFTWARE DEVELOPMENT UNCERTAINTY MAY ADD UP TO $15M, POSSIBLY OFFSET BY
BETTER NEGOTIATED PRICES ON HARDWARE, INCLUDED HERE AT LIST PRICE
SECRET
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
(Table V)
COST/FUNDING PROFILE
(IN MILLIONS OF 1982 DOLLARS)
SOFTWARE INTEGRATION OPTION (BUY/PURCHASE PLAN)
SOFTWARE
31.9*
CONVERSION
6.5
COMMUNICATIONS
.7
HARDWARE
-
COMPUTER EQUIP
75.0*
TERMINALS
13.1
-
PRINTERS
3.5
ADMINISTRATIVE
- DEV FACILITY
1.5
- RELOCATION
1.6
- CONTRACTOR TRAVEL
.5
- STAFF TRAVEL
.7
*SOFTWARE DEVELOPMENT UNCERTAINTY MAY ADD UP TO $15M, POSSIBLY OFFSET BY
BETTER NEGOTIATED PRICES ON HARDWARE, INCLUDED HERE AT LIST PRICE
rOECUT
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
(Table VI)
DIA/CIA 1984 PROGRAM SUBMISSION
($ MILLION)
4th Qtr
FY82 FY83 FY84 FY85 FY86 FY87 TOTAL
- DIA 3.0 12.8 13.0 17.1 14.4 10.7 71.0
Cr r) E:-r
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
 Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
DEVIL LOPMERT APPROAcn
--
-
3
141--
i I 'W
Ill
1 luV
Il t
1L.J~l14
INTEGRATION APPROACH
1218
17,0
14,5
8.8
6619
t 11MU11CHA6E PLAN)
08,
819
X612
14,0
8-2
58,1
01A/CIA :.:
3,0
12a
860
1
11.3
?
12
10
5 43
'OV6147 8001!-T
i7~1W~fMlgn
~ +
'r~
r ` '? T
wN
q //yy~~T 11
D~~~~? ? ! .J i !'~1 '{'tel.
;> ' 'FRb!T~ ':.,~(y~ ,1p11.#1F aY.~ }'y J 4
~:
FYPR
1
F
2) Lip
LI ~'O ,T ~.;,
of f t~.
, .~
~~!?. -
I
'
~?~
,~
~._ 1{',?
J
_
-.~IwIr7'
~( w+..rrn
i 47
' 9)~
?'
; .~
,41?
.w
y~' '~.
r
~;r
0C3
I
103
9-
J~
skr '^it .
P
YJ~
1 ? . A 1 E D - ,t ; .:; _....711 7
t ~1 ~VELO1 M NT;` `
UNO!RTAINTY MAY ADD $5M
~. SOFTWARE DEVELOPMENT
UNCERTAINTY. MAY ADO $ISM '
COST/FUNDUNO PRO ILS SUMMARY
QsJ1,t1i+~ c~ It_,,?,_ f wUt
111 1114 1 1'i 11114
1114W i Al
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
4 ,MILLION).
Y04' ' FY85 -
 Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1
Next 12 Page(s) In Document Denied
Approved For Release 2007/06/14: CIA-RDP84BOO148R000200440033-1