SICo P Special Briefing Federal Semantic Interoperability Community
SICo. P Special Briefing Federal Semantic Interoperability Community of Practice (SICo. P) August 9, 2007 (Updated August 13 and 23, 2007) For Mr. Michael Krieger Director Information Policy OSD, Do. D CIO 1
Overview n n n SICo. P provided a set of briefing slides for the August 9 th meeting (Sections 1 -4). SICo. P addressed the issues raised in the August 9 th briefing by supplementing the slides on August 13 th with notes (Section 5). SICo. P had an extensive email discussion which the SICo. P Co-chairs compiled and distilled in the Summary Points on August 23 rd (see next three slides). 2
Summary Points n n The leadership of the Do. D Co. I (Mike Todd) and SICo. P (Brand Niemann) worked together on the FEA/OMB DRM 2. 0 - The Do. D Co. I was featured as a best practice for information sharing in a Co. I and SICo. P led the DRM 2. 0 Implementation Through Testing and Iteration Work Group. The Do. D Co. I and SICo. P continue to interact in the Do. D Co. I Quarterly Meetings and through those with joint membership like Jim Schoening who leads the SICo. P Cross-Domain Semantic Interoperability WG (CDSI WG) that produced a white paper that was discussed in the press and gave rise to the August 9 th briefing for Mr. Krieger and his MITRE staff. 3
Summary Points n n Do. D is using semantic technologies and standards, and the recent Do. D Co. I Quarterly Meeting on July 31 st featured two presentations of that (David Hanz, SRI, and Mary Parmelle, MITRE). The SICo. P members participating in the August 9 th briefing came away with a range of impressions of the Do. D Co. I leadership from (1) Do. D and the IC are about 3 - 5 years behind where we are and we're pulling away fast, to (2) we need to take the time to understand their use case for semantic technology and focus our discussion on how semantic technology can be used to support their mission. All the SICo. P participants came away with the desire to work on how to "get Do. D leadership moving in the right direction" at the upcoming NCOIC Plenary and WG Meetings September 17 -21 st, and the Metatopia Conference, November 5 -7 th. 4
Summary Points n n The SICo. P and NCIOC SIF WG activities are about adding value to and reusing the Do. D and Do. D Co. I net-centric information sharing work, not about critcizing, disrupting, or replacing it - we are two communities trying to better understand each other and help one another to achieve a common purpose - semantic interoperability in information sharing. SICo. P would like to see the Do. D Co. I Leadership and MITRE staff review and comment on the individual SICo. P member presentations on August 9 th, and especially the white papers from GSA, and give SICo. P members the opportunity to present our work in the Do. D Co. I Quarterly meetings and/or invite the Do. D Co. I members to the SICo. P and SOA Co. P meetings. 5
Agenda n n n 1. Introductions: Host 2. Introductions: SICo. P 3. SICo. P: Brief Summaries 4. Discussion 5. Post-briefing Notes 6
1. Introductions: Host n n Michael Krieger Clay Robinson Frank Petroski, MITRE Glenda Hayes, MITRE 7
2. Introductions: SICo. P n n Brand Niemann, Senior Enterprise Architect, USEPA, and SICo. P Co. Chair Rick Murphy, Senior Enterprise Architect, GSA Mills Davis, Managing Director, Project 10 X, and SICo. P Co-Chair Lucian Russell, Private Consultant 8
2. Introductions: SICo. P n n n Denise Bedford, Enterprise Architect, World Bank, and Georgetown University, Kent State University, University of Tennessee Facility Mike Lang, Founder, Revelytix Todd Schneider, Raytheon, and NCOIC 9
3. SICo. P: Brief Summaries n n n n A. SICo. P Overview: Brand Niemann B. SICo. P White Paper 1 and GSA Activities: Rick Murphy C. SICo. P White Paper 2: Mills Davis D. SICo. P White Paper 3: Lucian Russell E. Framework for Achieving and Managing Interoperability: Denise Bedford F. Semantic Wiki and New OS/NII Project: Michael Lang G. NCIOC Semantic Interoperability WG: Todd Schneider 10
3. A. SICo. P Overview n n SICo. P was charted under the Best Practices Committee of the Federal CIO Council in March 2003 and has delivered three white papers and produced eleven conferences. SICo. P led the OMB/FEA DRM 2. 0 Implementation Team. SICo. P has given Special Recognitions (35) that document the progress along the Spectrum of Reasoning and Applications. SICo. P actively participates in Do. D Co. I, W 3 C, Semantic Technology, NCOIC, etc. work groups and conferences. 11
3. A. SICo. P Overview n SICo. P Has Three White Papers: ¡ Introducing Semantic Technologies and the Vision of the Semantic Web: n ¡ Semantic Wave 2006 - Executive Guide to the Business Value of Semantic Technologies: n ¡ W 3 C Semantic Web and DARPA DAML Program/SICo. P Semantic Web Applications for National Security (SWANS) Conference April 2005 (40 exhibits) 2006 Semantic Technology Conference. Updated at 2007 Conference. Operationalizing the Semantic Web/Semantic Technologies: A roadmap for agencies on how they can take advantage of semantic technologies and begin to develop Semantic Web implementations (recently released for public review): n Advanced Intelligence Community R&D Meets the Semantic Web (ARDA AQUAINT Program). 12
3. B. SICo. P White Paper 1 and GSA Activities Do. D Component and IC Member Extensions COI Extensions on m m Co ore og C EI E C Coomm re o n L n C 2 Common Core Universal Core IS R C Co om re mo n Business Common Core Lessons Learned in Core. Based Information Sharing (2004 -2007): ¡ After two rounds of investment in corebased approach … Assumptions that didn’t pan out for us … n Fact: XML-based technologies can’t represent semantics. Fact: OMG can’t field QVT to achieve core capabilities. Fact: Short supply of folks with adequate background in information theory. n n Adequate mainstream technology set available. Sustainable labor cost at scale. Manage consistency through convention. 13
3. B. SICo. P White Paper 1 and GSA Activities n Early Stage 1 – Language composition with description logics: ¡ ¡ ¡ 3 rd pilot (current), production (2 -5) years. Tomorrow’s tools are open source today. Sustainable core with manageable labor costs at scale: n n n Knowledge standardized and externalized v. XML/UML. Consistency and satisfy-ability (guaranteed) by construction through automated reasoning services. Extended unification and alignment (mapping) approach. ¡ Shared concept (remedial) v. categories. owl (semiotics) 14
3. B. SICo. P White Paper 1 and GSA Activities n Remaining stages …: ¡ Later stage 1 - (4 – 7) years: n n n ¡ IF-Map derived incremental unification and alignment. Testing Chu space transforms and end-to-end translation. Barriers: Knowledge transfer. Stage 2, 3 & 4 (6 – 20) years n n Composition and refinement of algebraic specifications. Additional reasoning services (unification, resolution, gmp). Non-monotonic logics and Institutions. Barriers: Retooling investment, Semantic Web resistance. 15
3. B. SICo. P White Paper 1 and GSA Activities n For more information …Please read our papers: ¡ ¡ ¡ n Roadmap for Semantics in Net-centric Enterprise Architecture Information Flow in the Federal Enterprise Redux: Governing Federations, Sharing Information and Ensuring Privacy Perspectives on a Unifying Framework for the Federal Enterprise … our stuff is open source, let’s share … 16
3. C. SICo. P White Paper 2 17
3. C. SICo. P White Paper 2 18
3. C. SICo. P White Paper 2 n Education: ¡ Metaland — Web 3. 0 semantic wiki magazine: n ¡ http: //metaland. visualknowledge. com/ Metatopia — DAMA NCR Chapter Conference, November 5 -7, 2007: n http: //www. wilshireconferences. com/metatopia/ 19
3. D. SICo. P White Paper 3 n n The Data Reference Model (DRM) Version 2. 0 had three components, Data Description, Data Context and Data Sharing. It pushed details to Communities of Interest because they “knew best”. It encapsulates but does not solve all pre 2006 problems with data representation and sharing. In 2006 several Computer Science breakthroughs occurred which made DRM 2. 0 obsolete: DRM 3. 0 will have only two sections. 20
3. D. SICo. P White Paper 3 New Computer Science Breakthroughs n n n English was ambiguous and could not be made otherwise Verbs describing processes could not be described Schemas, XML etc. were near useless n n n English has been disambiguated and exact descriptions can now be created Formal descriptions of Verbs’ processes are possible and unique Schemas can be made usable 21
3. D. SICo. P White Paper 3 n Failure Points Can Be Addressed: ¡ ¡ COIs work can be made useful provided the exact semantics of their work products are specified exactly Schemas need to be accompanied by exact English descriptions of their meaning, but if so can be useful Knowledge Bases can now be built …. But: you cannot use 1960 -70 s methods to do this, nor just XML! You also need better contractor support, some A-team members. 22
3. D. SICo. P White Paper 3 # 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Semantic Relation POSSESSION KINSHIP PROPERTY-ATTRIBUTE HOLDER AGENT TEMPORAL DEPICTION PART-WHOLE HYPONYMY ENTAIL CAUSE MAKE-PRODUCE INSTRUMENT LOCATION-SPACE PURPOSE SOURCE-FROM TOPIC MANNER MEANS ACCOMPANIMENT-COMPANION EXPERIENCER Abbr POS KIN PAH AGT TMP DPC PW ISA ENT CAU MAK INS LOC PRP SRC TPC MNR MNS ACC EXP # 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 Semantic Relation RECIPIENT FREQUENCY INFLUENCE ASSOCIATED-WITH / OTHER MEASURE SYNONYMY-NAME ANTONYMY PROBABILITY-OFEXISTENCE POSSIBILITY CERTAINTY THEME-PATIENT RESULT STIMULUS EXTENT PREDICATE BELIEF GOAL MEANING JUSTIFICATION EXPLANATION LCC’s Polaris Semantic Relations Abbr REC FRQ IFL OTH MEA SYN ANT PRB PSB CRT THM RSL STI EXT PRD BLF GOL MNG JST EXN 23
Cyc contains: 15, 000 Predicates 300, 000 Concepts 3, 200, 000 Assertions Represented in: • First Order Logic • Higher Order Logic • Modal Logic • Context Logic Micro -theories 3. D. SICo. P White Paper 3 Thing Intangible Individual Thing Sets Relations Space Physical Objects Living Things Ecology Natural Geography Political Geography Weather Earth & Solar System Human Beings Human Artifacts Human Anatomy & Physiology Partially Tangible Thing Time Events Scripts Artifacts Plans Goals Physical Agents Animals Mechanical Software Social Language Relations, & Electrical Literature Devices Works of Art Culture Organizational Actions Organizational Plans Agent Organizations Social Behavior Agents Actors Actions Movement State Change Dynamics Plants Temporal Thing Logic Math Borders Geometry Emotion Human Products Conceptual Perception Behavior & Devices Works Belief Actions Vehicles Buildings Weapons Paths Spatial Paths Materials Parts Statics Life Forms Spatial Thing Social Activities Human Activities Business & Commerce Purchasing Shopping Types of Organizations Politics Warfare Sports Recreation Entertainment Transportation & Logistics Human Organizations Nations Governments Geo-Politics Professions Occupations Travel Communication Law Everyday Living Business, Military Organizations General Knowledge about Various Domains Specific data, facts, and observations Cycorp © 2006 24
3. E. Framework for Achieving and Managing Interoperability n Meeting the Interoperability Challenge: ¡ Critical success factor is to get ahead of the harmonization problem n n ¡ Essential Components for Interoperability: n n n ¡ ¡ Stop chasing and start managing the problem. Lack of direction and guidance at all levels of architecture leads to lack of interoperability. Core metadata (Do. D has achieved this). Metadata semantics and specifications for each attribute. Guiding Principles at the attribute level. Participation of stakeholders. Defined governance roles and processes for attributes. Metadata and semantics provide the framework that supports participation. Participation, guiding principles and governance models provide the synchronization and agility that is necessary for a complex, dynamic and multi-disciplinary organization. 25
3. E. Framework for Achieving and Managing Interoperability n Degrees of Interoperability: ¡ First degree -- core metadata set: n n n ¡ Identify the essential core set from what exists. Map and harmonize what exists. Implement new attributes. Design metadata architecture around core metadata. Define and publish specifications. Second degree – manage values: n n n Define guiding principles, policies and governance roles for core metadata. Establish change management process for core and institutional metadata. Establish and synchronize roles and responsibilities across the organization – these can be flexible but bounded. Involve business owners, domain experts, ontologists, taxonomists and metadata managers in appropriate roles. Whether you use a federated or centralized governance model is less important than the lack/availability of a governance model. 26
3. E. Framework for Achieving and Managing Interoperability n Degrees of Interoperability (continued): ¡ Third Degree -- converge application and service development: n n ¡ Development has a metadata component – review is at the attribute level. Ensure that all COIs and systems developers are building to the same framework. Anchor development around an enterprise level metadata dictionary. Involves continuous awareness raising and advocacy. Fourth degree – automate and monitor values: n n Configure semantic technologies to support metadata. Embed ‘vocabularies’ into technologies. Shift Co. I domain expertise to building and populating the semantic tools. Leverage semantic technologies to support consistent capture and population of the metadata values. 27
3. E. Framework for Achieving and Managing Interoperability n Lessons Learned and Best Practices: ¡ ¡ ¡ Develop guiding principles and governance processes for Do. D core metadata attributes. Broaden the ‘ownership’ of governance by implementing a governance process for each core attribute. Continue to leverage the Co. I’s but clarify their roles in the governance model. Implement a common and automated method of generating ‘metacards’ across the agency. Ensure that you have a well defined and efficient metadata architecture at the enterprise level. Implement semantic technologies (with embedded integrated values) at the core metadata attribute level at the ‘metacard’ level. 28
3. F. Semantic Wiki and New OS/NII Project n Architecture Federation Pilot: ¡ ¡ Army Chief Information Officer (CIO/G-6) Architecture Operations, Networks and Space (AONS) Directorate, Army Architecture Integration Center (AAIC) 29
3. F. Semantic Wiki and New OS/NII Project n Purpose: ¡ Portfolio management and investment planning activities: n The Do. D EAC COP had previously determined that a complex set of diverse information sources need to be tapped into and merged with the assets represented in architectures in order to supportfolio management. 30
3. F. Semantic Wiki and New OS/NII Project n Purpose (continued): ¡ Integrate knowledge in support of investment decision making by the following stakeholders: n n Joint Net-Centric Operations Capability Portfolio Manager (JNO CPM). Joint Focused Logistics CPM. Joint Command & Control (JC 2) CPM IT Portfolio Managers at both Do. D and Service levels. The Business Transformation Agency (BTA). 31
3. F. Semantic Wiki and New OS/NII Project n Classification Ontology: ¡ Engage a COI to build an ontology that is suitable for classifying the diverse collection of artifacts: n n n Common service capabilities and their related definitions. An architectural model of a sub-set of currently deployed Army tactical enterprise services, developed in DODAF constructs. An architectural model of a sub-set of Army tactical enterprise services as planned for future deployment. An architectural model of a sub-set of supporting Do. D IT infrastructure. An EXCEL spreadsheet containing a set of Army systems and related attributes. 32
3. F. Semantic Wiki and New OS/NII Project n Semantic Wiki: ¡ A semantic wiki will be used to develop the classification ontology: n n n A Community of Interest will be formed on the wiki. Stakeholders will collaborate over the definition of the ontology. The ontology will be published to a repository. 33
3. F. Semantic Wiki and New OS/NII Project n Repository: ¡ ¡ n The ontology will be imported into a commercial repository. All of the architectural/capability artifacts will be imported into the repository. Semantic Mapping: ¡ A semantic mapping tool will be used to match all of the artifacts to the semantically correct nodes in the ontology. 34
3. F. Semantic Wiki and New OS/NII Project n Result: ¡ Having all of the artifacts related to the ontology will facilitate the answering of the following kinds of questions: n See next two slides. 35
3. F. Semantic Wiki and New OS/NII Project n n n What are the capabilities and capability increments of interest? What are the representative echelons of the federation: nodes and ”edges”? For each echelon: ¡ ¡ n What are the minimum and desired capabilities, and when are these needed? What are the points of intersection/interfaces between echelons? Who are the providers (POR or otherwise) and consumers? What are the federation alternatives for each point of interface? For each provider: ¡ ¡ ¡ What increment of a given capability does it offer and what capacity does it have to serve consumers by FY? How much will this service cost to procure by FY? How much will this service cost to sustain by FY ? 36
3. F. Semantic Wiki and New OS/NII Project n n n What dependencies does this service have on other providers by FY? What is the ’interface’ (e. g. protocol) provided for federation with other providers to implement a given enterprise-wide capability? For each ‘interface’: ¡ ¡ What is that interface’s dependence on lower-level protocols? What are the Qo. S/transport/bandwidth dependencies? What are the rules for connecting to it? (e. g. information assurance (IA) credentials? ) What is the quality/accuracy/timeliness/consistency specification for that interface (e. g. : ”we will provide blue force locations that are no more than 5 minutes old within nearest 1000 meters. ”) 37
3. G. NCO Industry Consortium (NCOIC) • Vision: Industry working together with our customers to provide a network centric environment where all classes of information systems interoperate by integrating existing and emerging open standards into a common evolving global framework that employs a common set of principles and processes. n Primary tenets of the Consortium’s vision: ¡ Work to identify and develop a Network Centric environment ¡ Provide assured technical interoperability ¡ Embrace, enhance, and encourage open standards ¡ Establish and educate on common principles and processes Mission: n Facilitate global realization of the benefit inherent in Network Centric Operations. We seek to enable continuously increasing levels of interoperability across the spectrum of joint, interagency, intergovernmental, and multinational industrial and commercial operations. We will execute this mission in good faith as a global organization with membership open to all enterprises in quest of applying the vast potential of network centric technology to the operational challenges 38 faced by our nations and their citizens.
3. G. NCOIC Interoperability Framework (NIF) n n n Basis for designing and building interoperable Network Centric Systems in diverse customer and supplier worldwide environment. Organizes and structures core principles of Network Centric Operations such as connectivity, interoperability, security, discovery, and end-to-end Quality of Service. Multiple working groups addressing different aspects of interoperability 39
3. G. NCIOC Semantic Interoperability (SIF) WG This wg will develop the NIF Semantic Interoperability (SIF) Framework, a sub -framework of the NCOIC NIF overarching framework. The SIF framework will include: n Development of an evolving technology and capability map of all the relevant concepts and technologies associated in this domain, n Development of a clear and unambiguous definition of semantic interoperability, n Identification and description of the impact that various levels of semantic interoperability have on networked operations, n Evolution of an overarching semantic interoperability model, n Identification and taxonomic classification of the semantic interoperability problems and examples of within various applications, n Development of semantic interoperability principles and tenets, n Identification and development of semantic interoperability architecture patterns. n Emerging technology concepts will also be discussed as to their scope of problem solution and possibly their maturity level and related standards. 40
3. G. NCOIC SIF Objectives n n Develop a comprehensive understanding of the problems of semantic interoperability and define a semantic interoperability framework (SIF) where the scope and role of each problem can be illustrated and where problem specific architectural pattern solutions can be integrated (Services, Situational, and Knowledge Sharing domains) Investigate and describe the use of semantic technologies and standards to create mutually consistent understanding shared of shared knowledge among different interoperable entities for different context situations Define the role of the semantic interoperability framework (SIF) in an NCO environment Develop NCO Capability Specific Semantic Interoperability Patterns 41
4. Discussion n Upcoming Opportunities: ¡ September 11 -12 th, SOA Institute’s SOA Conference: n ¡ Federal SOA Co. P (Federal Jump Start Kit: Open Source/IONA FUSE Technologies). October 1 -2 nd, 4 th SOA for E-Government Conference at MITRE, Mc. Lean, VA: n Federal SOA Co. P (Multiple demos including one of the FFATA use case). 42
4. Discussion n Upcoming Opportunities: ¡ September 17 -21 st, NCOIC Plenary Session and WG Sessions: n ¡ October 16 th, NIST Net-Ready Sensor Standards Harmonization WG Meeting: n ¡ Services Information Interoperability WG. SICo. P: Modular Approach to SSHWG Ontology and Demonstrations November 5 -7 th, Metatopia Conference: n Semantic Technology Track Being Organized 43
4. Discussion n Do a hands-on vocabulary harmonization session following the October 16 th COI Meeting? Do an all-day Collaborative Expedition Workshop? Do a SICo. P Special Conference? ¡ A Third Conference has been planned. 44
5. Post-briefing Notes n A. SICo. P Overview: Brand Niemann ¡ Issue: I don’t understand who SICo. P is and what it does? n Answer: I now understand who SICo. P is and think they do good work. Please followup with Clay Robinson. 45
5. Post-briefing Notes n B. SICo. P White Paper 1 and GSA Activities: Rick Murphy ¡ Issue: I don’t have a semantic technology problem, but a syntax governance problem. n Answer: Rick’s GSA group is 3 -5 years ahead in its thinking and work on this problem. 46
5. Post-briefing Notes n C. SICo. P White Paper 2: Mills Davis ¡ Issue: Was there one? n Answer: Our research and education activities can help you understand the broader picture. 47
5. Post-briefing Notes n D. SICo. P White Paper 3: Lucian Russell ¡ Issue: Registering XML Schemas in the Do. D Metadata Registry is all we need for semantic interoperability. n Answer: XML Schemas in a registry to do not solve the Semantic Interoperability problem. 48
5. Post-briefing Notes n E. Framework for Achieving and Managing Interoperability: Denise Bedford ¡ Issue: Was there one? n Answer: This is the most important part of the briefing to help you. 49
5. Post-briefing Notes n F. Semantic Wiki and New OS/NII Project: Michael Lang ¡ Issue: Who and what is this project for? n Answer: Mike Lang provided the details. 50
5. Post-briefing Notes n G. NCIOC Semantic Interoperability WG: Todd Schneider ¡ Issue: NCOIC is just for the large companies and has different problems than the Do. D Co. I. n Answer: the NCOIC SII WG has a cooperative agreement with SICo. P to bring in the small, innovative companies, and has similar problems to the Do. D Co. I and wants to followup to explain more fully. 51
5. Post-briefing Notes n General: ¡ Issue: We had a time management problem. n ¡ Answer: They had issues with most of what we presented and asked a lot of questions. Issue: The last two presentations at their Do. D Co. I meeting used ontologies and reasoning for semantic interoperability. n Answer: Senior Do. D management was unable to stay to hear them and the subsequent discussion. 52
- Slides: 52