Federal Government IT Strategy

1
Federal Government IT Strategy

• Michael Lang
• January 8, 2007

2
Background

• I founded Metamatrix eight years ago
• The federal government became our largest
  customer by accident
• I have worked with dozens of federal IT programs
  and with dozens of integrators
• Mostly interested in information management and
  systems architecture
• Now concentrating on semantic technology

3
Agenda

• Federal IT Overview
• Federal Enterprise Architecture
• Net Centric Enterprise Services
• Communities of Interest
• Domain Vocabularies
• Semantic Technology

4
Federal IT Investment

• Your Federal Government is doing billions of
  dollars of RD in the IT area
• There are hundreds of IT programs
• Orion NASA
• Trailblazer, Groundbreaker Ft Meade
• TTIC, US Visit DHS
• Sentinel, NDEX, RDEX DOJ
• DLA IDE, GCSS, GCCS - DOD

5
Federal IT Communities

• There are three distinct communities in the
  Federal IT space
• Intelligence
• Looks a little like financial service firms
• Department of Defense
• Looks most like commercial enterprises
• Civilian
• All three have very different use cases and
  agendas

6
Intelligence

• Pre 9-11 systems were all secure silos
• Sharing was avoided
• Security was paramount
• A lot of custom code
• Fair mix of structured and unstructured
  information
• Use case is analysis

7
Intelligence

• An Executive Order mandating information sharing
  across the intelligence community was issued
  right after 9-11.
• Information sharing is now paramount
• Metadata management is key
• Logical data models for each domain
• Data is being exposed as services
• Progress is very slow because of security concerns

8
Department of Defense

• Mission changed with the collapse of the Soviet
  Union and the arrival of Don Rumsfeld
• Much nimbler warfighter
• Smaller missions, faster response
• Requires better co-ordination between military
  branches and commands
• Largely client server
• Mostly structured information

9
Department of Defense

• Move to SOA is well under way
• Data being exposed as services
• Registries and repositories proliferate
• Many domain data models
• Many, many efforts under way to achieve greater
  degrees of interoperability
• Throw spaghetti at the wall and see what sticks

10
Civilian

• Mission changed with the arrival of the Internet
• Executive order creates eGov initiative
• Citizen centric services
• No sense of urgency here
• Relatively small budgets

11
FEA and NCES

• Federal Enterprise Architecture
• And
• Net Centric Enterprise Services

12
Overarching Programs

• There are two long running, overarching IT
  initiatives whose goal is to re-engineer the
  federal government IT infrastructure
• FEA, Federal Enterprise Architecture
• Managed by OMB
• Top down
• NCES, Net Centric Enterprise Services
• Managed by DOD, DISA
• Bottom up

13
FEA

• This program began in 2002 as a result of an
  executive order from the White House that created
  the eGov initiative
• http//www.whitehouse.gov/omb/egov/
• To transform the Federal government to one that
  is citizen-centered, results-oriented, and
  market-based, the Office of Management and Budget
  (OMB) is developing the Federal Enterprise
  Architecture (FEA), a business-based framework
  for government-wide improvement.

14
Architecture Principles FEAPMO

• Motherhood and Apple Pie
• The federal government focuses on citizens
• The federal government is a single, unified
  enterprise
• Federal agencies collaborate with other
  governments and people
• Information is a national asset
• The federal architecture is mission-driven
• Security, privacy and protecting information are
  core government needs
• The federal architecture simplifies government
  operations

15
FEA Reference Models
16
FEA Current State

• Even though there are budgetary enforcement
  procedures mandating agencies to begin
  implementation of the FEA, they are largely
  ignored
• The root of the problem is that the architecture
  does not hang together and the prospective users
  know it
• The DRM is not credible

17
Data Reference Model

• I spent two years working on the DRM, it is the
  most troublesome layer of the stack
• The DRM provides a standard means by which data
  may be described, categorized, and shared. These
  are reflected within each of the DRMs three
  standardization areas
• Data Description Provides a means to uniformly
  describe data, thereby supporting its discovery
  and sharing
• Data Context Facilitates discovery of data
  through an approach to the categorization of data
  according to taxonomies additionally, enables
  the definition of authoritative data assets
  within a community of interest (COI)
• Data Sharing Supports the access and exchange of
  data where access consists of ad-hoc requests
  (such as a query of a data asset), and exchange
  consists of fixed, re-occurring transactions
  between parties

18
NCES

• Net Centric Enterprise Services
• NCES started at about the same time as FEA, but
  is an initiative out of DISA (Defense Information
  Systems Agency) the CTO office of DOD.
• NCES does not pay much attention to FEA
• Global Information Grid GIG
• Includes the physical networks and other hardware

19
NCES Mission

• NCES will enable the secure, agile, robust,
  dependable, interoperable data-sharing
  environment for DOD where warfighter, business,
  and intelligence users share knowledge on a
  global network. This, in turn, facilitates
  information superiority, accelerates
  decision-making, effective operations and
  net-centric transformation.
• To enable successful conduct of warfare and other
  operations in the Information Age.
• Make information available on a network that
  people can depend upon and trust.
• Populate the DOD networks with new, dynamic
  sources of information to defeat the enemy.
• Sounds a lot like any commercial enterprise
  mission statement

20
NCES Mission

• NCES represents a different approach to building
  and fielding DOD Information Systems
• Market-based approach, recognizing that a user's
  information technology (IT) needs are dynamic and
  are rarely satisfied by systems built with a set
  of pre-determined user needs
• Users themselves are best able to define their
  requirements
• The NCES approach is DOD-wide
• It offers unprecedented access to information
  from global sources while leveraging existing IT
  investments

21
NCES Current State

• Service Oriented Architecture
• A lot of the infrastructure is in place
• Metadata catalogs/repositories
• Services Registry
• Tools for converting relational to XML
• Tools for creating and publishing services
• XML Schemas describing domains
• Quality of service software
• Security software and hardware
• Governance

22
NCES Current Bottleneck

• Interoperability
• As soon as the number of services proliferate
• The number of silos proliferate
• They are more granular but still hard to use and
  manage
• Pulled a lot of the funding from programs that
  are creating services
• Funding a lot of pilot projects to solve
  interoperability

23
Domain Vocabularies

• Early efforts used XML Schema and ER diagrams to
  define the domain data model
• Global Justice XSD
• National Information Exchange Model NIEM
• Command and Control C2IEDM
• Not extensible, not semantic
• No connection between the businessperson and the
  data

24
Communities of Interest

• Communities of Interest form to create domain
  vocabularies
• All of the terms in a domain
• Data dictionary, logical model, schema
• What they mean
• How they are used
• How they are related
• The Domain vocabulary is the interoperability
  master key
• All data elements in all systems are mapped to
  terms in the domain vocabularies

25
Use of Vocabularies

• Permit humans express their concepts in a machine
  readable language
• Enable machines to perform the data translation
  and transformation required by data integration
• Vocabularies are the essential underpins to
  sharing data or system interoperability that
  requires dynamic links among unknown, unlimited
  numbers of data sources
• Essential to all semantic technologies, including
  semantic search

26
Semantics

• Most programs have moved to OWL for defining
  domain vocabularies
• http//www.opengroup.org/projects/soa-ontology/
• http//osera.gov/web/guest/projects/fea-rmo
• Flexible and extensible
• Naturally distributed, URI and URLs
• Best design-time metadata representation model
• Machine readable at runtime
• Functions at the scale of the WWW

27
Semantic Technology Standards
OWL Ontology
W3C Semantic Technology Standards
28
Solving Data Relationships (Related)
Why Ontologies are so important An ontology is
an abstract representation of concepts and their
relationships that enables deductive and
inferential reasoning upon itself. They are
uniquely capable of creating relationships,
otherwise impossible to identify on a mass scale,
that explicitly reason for all relationships.
29
MBIs SOA-Enabled DoDIIS Data Layer

• Use Ontology to semantically match elements
  across disparate sources
• Build virtual layer
• Service enable data layer

30
Government Leads the Way

• Semantic technology
• The government last led the charge with
  relational database technology and IP networks
• DARPA funded the RD for RDBMS for 10 years
• And then became the early adopter
• DARPA created OWL (DAMLOIL) eight years ago
• Numerous projects funded to employ semantic
  technology
• Just making it into operational systems

31
Conclusions

• Bottom up architectural approach works better
  than top down
• Communities will form and participate in the
  construction of the system especially the domain
  vocabularies
• The effort should and can include business
  people, technology people and data people

32
Conclusions

• For transactional systems, data is being
  represented by XML and exposed as services (WSDL)
  in an SOA
• Domain vocabulary is being described in OWL
• Interoperability
• For analysis, data is being represented as RDF
  and queried using SPARQL
• The ontology is the integration layer

33
Thank You

• Michael Lang
• michaelalang_at_gmail.com

34
Discovering and Binding Services
You can haveone or moreof these
Mapping Vocabularies A B
Mapping Vocabulary
same as or same class as
Vocabulary A
Vocabulary B
generate
describe the RDF
generate
describe the RDF
XML Messages (in RDF XML)
XSD
XSD
XML Messages (in RDF XML)
Describe the structure (elements attributes)
Describe the structure (elements attributes)
reference
reference
WSDL
WSDL
describe
describe
35
Using Service Responses
KNOWN FACTS
36
Semantic Interpreter or Semantic Message
Translator
Small wrapper around Jena
Vocabularies (OWL) Composed at design-time
submit
produce
QUERY (SPARQL)
KNOWN FACTS
NEXT SERVICE REQUEST MESSAGE
Designed to obtaindesired messagefor next
service call
Composed from previousmessages in a
SOAtransaction plus assertions(facts) obtained
fromother sources
37
Single Vocabulary/Dictionary
Composite(s)
Unit Identification Code
Nationality
Fields
Armed Service
Sequential Location Number
Valid Entries
Nationalitystring(2) - enumeration value"AF" -
enumeration value"AL" - enumeration value"AG" -
enumeration value"AQ" - enumeration value"AN"

Armed Servicestring(1) - enumeration
value"F"/gt - enumeration value"A"/gt -
enumeration value"C"/gt - enumeration
value"B"/gt - enumeration value"J"/gt
Sequential Location Numberinteger - min
value"0000" - max value"999999" - pattern
value"0-94,6"
Other Metadata
38
Enterprise Vocabulary/Dictionary
Enterprise Vocabulary
Fields Composites Valid Entries
A
USMTF Vocabulary
Link 16 Vocabulary
VMF Vocabulary
B
D
A
Fields Composites Valid Entries
Fields Composites Valid Entries
Fields Composites Valid Entries
39
Logical (Relationship) View

• Reference Model for naming conventions,
  data-typing conventions, and business component
  structure
• Purely Conceptual -- Represents abstract view of
  data relationships within a vocabulary (cannot be
  queried from data)
• Improves ability to manage change and support new
  virtual models more quickly

40
Info Exchanges/Use Cases
Enterprise Vocabulary
Harmonized Standard Views
Sets Messages Web Services
A
Fields Composites Valid Entries
USMTF Vocabulary
Link 16 Vocabulary
VMF Vocabulary
Community Specific
B
D
A
Fields Composites Valid Entries
Fields Composites Valid Entries
Fields Composites Valid Entries
Specific Information Exchanges (Messages/Virtual
Models)