Position papers are due March 18, 2005 for the W3C Workshop on Rule Languages for Interoperability, to be held April 27-28, 2005 in Washington, D.C., USA. Hosted by ILOG S.A. with support from DARPA through the DAML program, this W3C Workshop is intended to "gather various participants and inputs needed to see how a standard rule framework might be developed, as informed by the Web Architecture and useful for addressing real user challenges."

The Workshop on Rule Languages for Interoperability is open to the public and there will be no participation fee. Interested participants should submit a position paper on a topic within scope for the workshop, with focus upon the requirements for the public representation and interchange of rules. "Position papers discussing interchange formats are expected to focus on the requirements and types of application covered by the proposal; papers discussing specifications or rule interchange format are expected to focus on that aspect and on how they could link to/import rules represented in other existing or emerging formats; papers discussing general issues regarding rules interchange and rule systems interoperability are expected to focus on how relevant existing standards or proposal or parts of an approach can be reused, evolved, extended; on principles and architecture; on related efforts in other communities (e.g., OMG, JCP, ISO, RuleML, SWSI, WSMO)." Accepted papers will also be made available to the public from the workshop web site.

The Workshop CFP reports that rule languages and rule systems "are widely used in applications ranging from database integration, service provisioning, and business process management to loan underwriting, privacy policies and Web services composition. General purpose rule languages remain relatively unstandardized, however, and rule systems from different suppliers are rarely interoperable."

In the W3C context, Semantic Web languages like RDF and OWL "are beginning to support data/knowledge sharing on the same scale and with considerable flexibility. Having a language for sharing rules is often seen as the next step in promoting data exchange on the Web."

The W3C Workshop announcement notes that "there is a thriving commercial market in several families of rule technologies, including production rules, event-condition-action rules, Prolog, relational database systems, and others. Business rules, decision tables, and decision trees are used to automate the enforcement of business policies and regulations. Logical formulas, constraints, ontologies, association and transformation rules are used for inferencing in information retrieval and information integration, including databases, and metadata repositories (e.g., Dublin Core Initiative), or in analytical, forecasting and/or optimization applications. However, practical interoperability between current systems, especially across the different families, is currently quite limited."

The goal of the Workshop on Rule Languages for Interoperability is to "gather and refine use cases and requirements for a framework, together with information about available technologies and relevant areas of practice and research. The effort will help establish a common ground for this work as well as a community of possible participants." Key deliverables are to include use cases accompanied by test cases, potential requirements for new work, identification of candidate technologies, published workshop position papers, workshop presentations, workshop minutes, and recommendations regarding future work.

The Workshop organizers envision that this public event may be one "step along the path to establishing a standard language framework to support rule system interoperation on the Web. It aims at gathering vendors, technologists, application developers and users to discuss and provide recommendations to the W3C regarding what is the best approach to the specification of a standard or family of standards for the public representation and exchange of rules on the Web, in terms of avoiding redundant efforts, of optimizing the potential for wide adoption, and of promoting consistency and interoperability between different applications or layers, while preserving their specific requirements."

Overview of the Workshop on Rule Languages for Interoperability

Background to the Rule Languages Workshop

Rules are everywhere. They are found in many domains, disciplines, and industries. Business policies, laws and regulations, guidelines and best practices, definitions and axioms, database schema translations, workflow branching and technical constraints, all require a declarative and modular approach to their implementation. There is a thriving commercial market in several families of rule technologies, including production rules, event-condition-action rules, Prolog, relational database systems, and others. However, practical interoperability between these systems, especially across the different families, is currently quite limited.

Rules are a key element of the Semantic Web vision, allowing integration, derivation, and transformation of data from multiple sources in a distributed, transparent, and scalable manner. Rules can themselves be treated as data, published on the web, and when URIs (or QNames) are used as symbol-constants in a rule language, they can form useful links between knowledge bases. In a Web services environment, rules offer the opportunity to enable the automation of the enforcement and composition of policies governing the delivery of information, the access to services, or the execution of processes.

Rules have advantages of flexibility and manageability. In addition, the declarative nature of rules gives them a special appeal as a programmatic and knowledge representation device in a distributed and Web-based environment, where they can be owned, specified and managed in one place, and applied in many other places. This requires, however, a standard way to represent rules unambiguously for publication and interchange purposes. Different rules and a common foundation.

Rules come in a variety of forms for different uses and applications. Business rules, decision tables, and decision trees are used to automate the enforcement of business policies and regulations. Logical formulas, constraints, ontologies, association and transformation rules are used for inferencing in information retrieval and information integration, including databases, and metadata repositories (e.g., Dublin Core Initiative), or in analytical, forecasting and/or optimization applications.

Rules, however, trace their roots back to formal logic. There, semantics can be represented via a logical model theory and inference can be based on logical proof theory. The most important de facto semantic standard is first order predicate calculus, unchanged for nearly a hundred years. In the last three decades, declarative logic programs have emerged as a complement to first order logic, and provided the foundation for the semantics of relational databases and many rule languages. Algorithmic techniques and theory for formal logic have been extended to enable, and semantically treat: procedural attachments for built-ins, tests, and actions; and non-monotonicity for negation-as-failure, defaults, inheritance, prioritization, updating, revision, and conflict handling...

To be effective, practical, and deployable, a Web standard on rules needs to focus on the requirements of end users and the needs of rule technology providers. The goal of being able to transfer rulebases / knowledge bases, or simply to process them with different software, has helped motivate several important standardization or standards-proposing efforts including RuleML and SWRL, n3, Metalog, KIF and ISO Common Logic, ISO Prolog, and others. Some of those have been aimed at more or less specialized purposes, e.g., in the domains of Web Service policies (WS-Policy, WSPL, Policy RuleML, SWSL, WSML), access control and authorization (XACML, EPAL, P3P/APPEL), Business Rules (BRML, SRML), and other areas as well. Related standardization efforts have also started with respect to rule modeling (OMG's Business Semantics for Business Rules RFP and Production Rules Representation RFP) and rulebase execution (JSR 94 - Java API for Rules Engine)..." [from the Workshop Call for Participation]

Workshop Organization

Workshop Chairs

• Sandro Hawke (W3C)
• Christian de Sainte Marie (ILOG)
• Said Tabet (The RuleML Initiative)

Program Committee

• Harold Boley (NRC Canada, RuleML)
• Dan Connolly (W3C)
• Mike Dean (BBN, DAML)
• Stefan Decker (DERI)
• Marc Goodner (SAP)
• Benjamin Grosof (MIT Sloan, RuleML)
• Pat Hayes (IHMC, Common Logic)
• Jim Hendler (University of Maryland)
• Ian Horrocks (University of Manchester)
• Sridhar Iyengar (IBM)
• Massimo Marchiori (University of Venice)
• Deborah McGuinness (Stanford KSL)
• Bob McWhirter (OpenXource, Drools)
• Eric Miller (W3C)
• Jon Pellant (Pegasystems)
• Jos de Roo (Agfa)
• Chris Swan (Credit Suisse First Boston)
• Paul Vincent (Fair Isaac)
• Note: list above current as of 2005-02-21

About ILOG S.A.

"ILOG delivers software and services that empower customers to make better decisions faster and manage change and complexity. Over 2,000 global corporations and more than 400 leading software vendors rely on ILOG's market-leading business rule management system (BRMS), optimization and visualization software components, to achieve dramatic returns on investment, create market-defining products and services, and sharpen their competitive edge. The BRMS market share leader, ILOG was founded in 1987 and employs more than 600 people worldwide..."