Alexander T. Borgida - From Software Specification to Requirements Specification Languages (and what might this have to do with Knowledge Representation)

22:00
Mercredi
2
Mai
2012
Organisé par : 

L’équipe "Keynotes" du LIG

Intervenant : 

Alexander T. Borgida, Dept. of Computer Science, Rutgers University, New Brunswick, NJ, USA


Alex Borgida is a Professor of Computer Science at Rutgers University, and is affiliated with the Cognitive Science Center there. He joined Rutgers after receiving his PhD from the University of Toronto (and spending a year backpacking around the world). He has continued to have a close association with the University of Toronto (where he holds an appointment that allows him to advise graduate students), and over the years has also had long-term research relationships with AT&T Bell Labs, and the University of Trento.

His research interest is in Conceptual Modeling, as applied to Databases, Knowledge Representation, and Software Engineering, with specific concern for languages, their logical semantics, and supporting tools. Highlights of his professional contributions include the ICSE "Most Influential Paper Prize" for work he co-authored on requirements modeling languages, and the co-design and implementation of Classic, the first Description Logic that saw serious industrial use ("helped configure billions of dollars’ worth of switching equipment").

The foundational work of figures such as Doug Ross and Michael Jackson has revealed the importance of focusing on the environment (Where ? Why ?) of the proposed software, not just its funtional specification (What ?). Significant phenomena accompanying this have been the rise of conceptual modeling and model structuring mechanisms, and especially the advent of Goal Oriented Requirements Engineering (GORE).

The talk considers some interesting trends in the languages/notations used in solving the requirements problem, including : from informal to highly expressive formal to less expressive formal notations ; the increasingly refined ontology of notions used in RE ; and the increasing significance of accommodating inconsistency. To add spice to the talk we consider interesting connections of our work to research in Knowledge Representation. (Joint work with John Mylopoulos et al).