Thoughts on Hosting an Ontology and Vocabulary Repository

Thoughts on Hosting an Ontology and Vocabulary Repository at OMG Evan K. Wallace – co-Chair OMG Ontology PSIG Manufacturing Systems Integration Division NIST 1

OMG’s Mission Since 1989 Develop an architecture, using appropriate technology, for modeling & distributed application integration, guaranteeing: n n n reusability of components interoperability & portability basis in commercially available software Specifications freely available Implementations exist Member-controlled not-for-profit 2

Who Are OMG? Accenture CSC HSBC Oracle Appian Corp. DND Canada IBM Penn National BAE Systems EDS IDS Scheer Sandpiper Software BEA Systems Fire. Star John Deere SPARX Blue. Phoenix Fujitsu Know Gravity SAP Boeing General Dynamics Lombardi Tele. Logic CA GSA Met. Life Tibco Capgemini HP NIST Vangent Citi. Group Harris Northrup Grumman VISA Cordys Hitachi No Magic W 3 C 3

A sampling of OMG specifications Common Object Request Broker Architecture - CORBA® Unified Modeling Language - UMLTM Common Warehouse Metamodel – CWMTM Information Management Metamodel – IMM (in process) Meta-Object Facility - MOFTM XML Metadata Interchange – XMITM Business Process Modeling Notation – BPMN Production Rule Representation - PRR Semantics for Business Vocabularies and Rules - SBVR Ontology Definition Metamodel – ODM 4

need for utility ontologies We are increasingly asked for “foundational”, “core”, or “utility” ontologies representing n Various IEEE, ANSI, ISO, & other standard vocabularies, both general (e. g. , currencies) & domain-specific (ACORD standards for Property & Casualty Insurance business vocabulary) n Metamodels, conceptual as well as logical, related to OMG standards (for SOA, event-driven architectures) n Where multiple artifacts are provided in the same repository w Ontology Definition Metamodel (http: //www. omg. org/cgibin/doc? ptc/2007 -09 -09) ontology & vocabulary formats: ODM/XMI for RDF, OWL, CL, Topic Maps with native forms w Semantics for Business Vocabularies & Rules (SBVR) equivalents, http: //www. omg. org/cgi-bin/doc? dtc/2007 -06 -06 w Mappings to production rules standards, such that the same vocabulary can be reused for rule generation: PRR (Production Rules Representation), http: //www. omg. org/cgibin/doc? bmi/07 -06 -06 slide provided by Sandpiper Software 5

reuse challenges vary Ontologies developed for programs such as the DARPA DAML program are aging n n n Ontology pages have not been revised since 2004 (see http: //www. daml. org/ontologies/) Most recent submission was actually in 2003 (see http: //www. daml. org/ontologies/submission. html) Community knowledge about development methodology & facts about the world relevant to the IC community have continued to evolve Many ontologies are developed for a specific purpose: domain or application oriented, but development assumptions that could impact reuse are not made explicit Research ontologies tend to be focused on demonstrationrelated content and are by nature incomplete, with varying coverage and levels of granularity due to funding limitations More recent ontologies available via the Protégé library are better documented, but many are also domain specific slide provided by Sandpiper Software 6

Potential OMG conceptual repository content For SBVR Foundational Vocabularies: Time and date Monetary amount Location Units of measure Quantities, cardinalities and ratios Arithmetic operations Collection operations To support ontology driven applications: ISO 3166 (country codes, mapped to language codes) ISO 639 (language codes) standard addresses (managed by the Universal Postal Union) ISO 4217 (currency codes, mapped to 3166) ISO 1087 (vocabulary for representing terminology) Various metadata standards such as ISO 5127, 11179 part 3, SKOS, Dublin Core 7

Possible processes for determining repository content Existing processes for adopting standards at OMG n n Request for Proposal (RFP) Request for Comment (RFC) Alternative mechanisms Submit what you want n Submit private content Should these include automated assessment and/or human evaluation components? n 8

RFP based Adoption Process Flow Business Comm. Recommendation Board Approval TC Recommendation Finalization and Revision Responses from Industry Final Submission TF -> AB Initial Submission Evaluation (by TF) and Revision Request for Proposal TF -> AB -> TC Letters of Intent 9

OMG specific challenges implementation requirement bicameral structure (domain and platform split) linking among models built from different metamodels and for different Communities of Practice (Business modelers versus Knowledge engineers) current organizational processes and staff not well suited to support continuing maintenance 10

General challenges Intellectual Property concerns particularly w. r. t. content based on International Standards Insuring availability and persistence maintenance and refreshment of content n need long term resource commitment what processes can be used how will freshness be monitored 11

Our approach Draft and issue an RFP for a well scoped content component to test the system and discover issues and practices Draft and issue an RFI to learn from others about the state of tooling, practices, and standards for repositories for knowledge models 12

Date-Time Foundational Vocabulary RFP http: //www. omg. org/cgi-bin/doc? bmi/08 -0302 Requests a consistent set of models for Date and Time concepts in the following forms: n n n an SBVR business vocabulary an Ontology based on one or more of the logic language metamodels in ODM (CL, OWL, RDFS) a UML class model A submission team is already forming, but one needs to be (or partner with) an OMG member to participate 13

Ontology and Vocabulary Management Information RFI http: //www. omg. org/cgi-bin/doc? ontology/08 -03 -02 Purpose: to guide the OMG on how to proceed in this area Anyone who wishes may respond Emphasis on 3 issues in content management n n n provenance – where the information comes from effectivity – at what time, location, and/or use is the content applicable or valid evolution – how we track change 14

Ontology and Vocabulary Management Information RFI Specific areas of knowledge requested: n n n n Application experience Tooling Tool interoperability Querying and accessibility Knowledge management and mapping Standards of practice Related or competing standards activities Example repositories 15

Contacts Evan K. Wallace -- co-Chair Ontology PSIG n ewallace@nist. gov Elisa Kendall -- co-Chair Ontology PSIG n ekendall@sandsoft. com Mark Linehan -- Chair Date-Time RFP submission team n mlinehan@us. ibm. com 16