User Assistance for Spreadsheets; A Case Study for Semantic Illustration

Kurzinhalt:

Spreadsheets are heavily employed in administration,  financial forecasting, education, and science because of their intuitive, flexible, and direct approach to computation. However, their success and growing size/complexity also  brings problems with usability and software engineering. We present an approach to alleviate these problems based on an explicit representation of the background knowledge associated with a  spreadsheet. We show how a theory-structured background ontology can be used to provide user assistance  at various levels by employing methods from mathematical knowledge management. The SACHS system implements this approach to provide a  semantic help system for DCS, an Excel-based financial controlling system.  The background  ontology is used in SACHS in two ways:  to provide semantic added-value services, and to semantically enhance the user interaction. In the final part of the talk we will abstract from the concrete example of spreadsheets to a new paradigm for semantic technologies. Instead of enhancing web resources into semi-formal ontologies by annotating them with formal objects that  allow reasoning as in the ``Semantic Web'' paradigm, the  ``Semantic Illustration'' architecture illustrates a software artifact with a semi-formal ontology by complementing it with enough information to render new semantic services in a ``Semantic Ally System''. We identify the property of ``semantic transparency'' as a user interface prerequisite  of Semantic Illustration and discuss it in the context of office applications.