PROCEDURES AND EXPERIENCES WITH THE ANNOTATION OF SERVICES AND RESOURCES TO AID IN GENERATION OF SERVICE WORKFLOWS

The trend toward service-oriented architectures has resulted in the development of several service endpoints within the geoinformatics community. Chaining of these endpoints into workflows, however, has been hindered by the lack of relevant domain-science metadata and semantics directly associated with the services and easily reviewable by both humans and computers. Standard Web Services Description Language (WSDL) documents only describe primitive types of data for input and output matching. This presentation will describe methods to associate these primitive data types with domain-relevant units and categories. We will demonstrate the inclusion of embedded Resource Description Framework (RDF) and eXtensible Markup Language (XML) Microformat nodes into the Document Object Model of Representational State Transfer (REST) -based services for both document- and procedure-centric services. We will also demonstrate the automated extraction of this information from these REST services using the W3C Gleaning Resource Descriptions from Dialects of Languages (GRDDL). The results are then transformed into XHTML for human review and automated indexing, and also into RDF collections that can be parsed and viewed through the use of common RDF tools. The overall system is a process that describes domain objects holding information and their relation and meaning, the composition of services with these descriptions, and the retrieval and viewing of this information by both human users and computer systems.

These methods are designed to aid in the description of services, both resources (data) and procedures (computational and transformational) to facilitate the creation of workflows in both manual and automatic methods. The NSF-funded project, CHRONOS, is joining in the development of AEON, a framework for automatically composing scientific workflows for the geosciences, by providing networked datasets and procedure services augmented by the description process needed for automated workflow generation. In addition, CHRONOS acts as a liaison between this effort and the geoscience community for field-testing.