The Goblin Project: Position Paper
Abstract:
This paper summarizes my views on the following issues: (1) the inappropriateness of a common meta-model which “integrates ” heterogeneous data sources versus an approach based on multiple conceptualizations, (2) the drawbacks of a single common ontology compared to an infrastructure in which many conceptualizations may co-exist, (3) the integration of problem-solving components into the information infrastructure, (4) the role and scope of mediation, (5) the scope of a domain, and (6) some of my views on what the architecture implemented in the GOBLIN project should look like. This is a collection of three independent discussions (each one occupying one section), rather than an integrated paper. 1 Critique of a Common Detector Model for CMS In previous sessions of our working group on the requirements of information integration at CERN, an approach to information integration that is based on the notion of a common meta-model (the Detector Description Database) has been introduced. This section is meant to make a point against the argument that a Detector Description Database as a single description of all CMS-related data is viable. My
