Authors: Said Tabet of RuleML: stabet@comcast.net Gert Wagner of RuleML: G.R.Wagner@tm.tue.nl Silvie Spreeuwenberg of LibRT: silvie@librt.com Paul Vincent of Fair Isaac Blaze Advisor: paulvincent@fairisaac.com Gonzaques Jacques of ILOG JRules: gjacques@ilog.fr Christian de Sainte Marie of ILOG: csma@ilog.fr Jon Pellant of Pega Systems: jon.pellant@pega.com Jim Frank of IBM: joachim_frank@us.ibm.com Jacques Durand of Fujitsu: mmjdurand@us.fujitsu.com
Accelerated by the vision of the semantic web, semantic technologies have recently made significant advances. The underlying methods and paradigms are already being transferred to adjacent areas of research in artificial intelligence, knowledge management, and elsewhere. Textbooks explaining the foundations have appeared. Large national and international projects on the topic are under way. The technology received a crucial impulse when the Web Ontology Language OWL became a W3C recommendation in 2004. It is already being perceived as a basic knowledge representation language with potential which goes far beyond the semantic web use case. At the same time, it is apparent that OWL needs to be extended with additional expressive features in order to become applicable and useful in many domains. Corresponding efforts are being pursued with frenzy by research institutions and industry. These efforts are being aided substantially by the fact that OWL is based on sound logical foundations as it can be understood as a well-understood decidable albeit very expressive fragment of first order logic. The need for an extension of OWL by rules has been known since the beginning. Indeed, the rule-based ontology representation language F-Logic [6, 1] is being used widely where a rule-based approach appears to be more feasible. As an example, we mention the HALO project1 by Vulcan Inc.2 whose ultimate goal is the creation of a “digital Aristotle”, an expert tutor in a wide variety of subjects, with deep reasoning abilities. The initial
This position paper is an attempt to enumerate and substantiate a number of key features, which, we believe, should be part of an expressive rule-based language to support the Semantic Web. This list of desiderata is based on our experience with the implementation and use of the FLORA-2 system [11, 31] as well as on our work on the languages for Semantic Web Services as part of the SWSL (http://www.daml.org/services/swsl) and WSMO (http://www.wsmo.org) projects.
Our brain’s solution to deal with this complexity was to create simplified reflections of reality in our minds with just enough details to allow us to process and understand our world. For example, I know that when I throw a stone in the air, it will eventually fall down and, probably, hit something. That’s enough information for me to conclude that throwing stones is not a good idea. I don’t need to know more facts about the stone such as its mass and structure, and I don’t need to be aware of the laws of physics that control the stone movement in the air. In software, we use a similar approach: we create XML representations of the real world with just enough details for the task at hand. For example, if we’re building a medical information system, we represent a person by an XML document that contains the person’s medical history. And if we’re building mortgage processing software, we represent the same person with an XML document that contains his/her financial information, but no medical details. Each XML document is a different reflection of the real person.
Logic based systems typically use a collection of Condition -> Action rules to encode the behaviour and response of the system to a specific set of events. These systems function by having a working memory of facts. By adding, removing or changing facts in the working memory, the rules system can see which (if any) rules need triggering. If the effect of a rule changes a state in the working memory then rules will be re-evaluated.
Northface University emphasizes model-driven development in its teaching and research programs as a key technology for empowering software engineering, is an active contributor to the Business Rules Team submission to the OMG’s RFP for adding a business semantics for business rules layer, and is currently developing an open source modeling tool suite based largely on ORM 2 (the next generation of ObjectRole Modeling), which is intended to be interoperable with a wide range of other rules-based tools. As such, we are interested in efforts to standardize languages for business rules to facilitate: • Rule declaration that can be readily understood and validated by non-technical domain experts; • Rule enforcement (automated, semi-automated, and manual); • Rule reuse and exchange with other tools, especially over the web.
Declarative Business Rules provides a framework for making software components flexible. Since several key business process activities are implemented using software components, one can argue that business rules technology can help in achieving business agility. Business rules raise the level of abstraction of specifications, making it possible for a business analyst to specify the rules which can be automatically translated to executable code. However, based upon our considerable experience in using business rules technologies for developing mission-critical Telecom Operation Support systems, we believe that major interoperability issues need to be addressed before the ‘ondemand’ flexibility vision in Information Systems can be realized. The position paper highlights some of those issues and suggests some approaches.
The Web Service Modeling Language WSML provides a framework for the modeling of ontologies and semantic Web services based on the conceptual model of the Web Service Modeling Ontology. In this paper we describe the two rule-based WSML-variants and outline our position with respect to a rule language for the Semantic Web. The first rule-based WSML variant, WSML-Flight, semantically corresponds to the Datalog fragment of F-Logic, extended with inequality in the body and locally stratified negation under the Perfect model semantics. The second, WSML-Rule, is an extension of WSML-Flight to the logic programming subset of F-Logic which allows the use of function symbols and unsafe rules (i
