Keynote Speakers

In a Ubiquitous World Requirements are Ubiquitous Too

Cecilia Mascolo, University of Cambridge
Cecilia Mascolo

Abstract: The soaring presence of devices that can sense the environment, human activity and social interactions in a ubiquitous fashion, opens the doors to potentially very effective multi-disciplinary research. Battery-powered tiny sensors can be distributed across an area to monitor conditions with very fine granularity. Moreover, mobile phones are powerful sensors that we voluntarily carry throughout our daily life. However, as well as introducing exciting opportunities, these technologies offer many challenges: writing software for these systems is all but obvious due to power, communication and computational constraints as well as to their context dynamicity. In addition, the interactions with the software users (i.e., the non computer scientists “scientists” collaborating in the projects) impose requirements that change the way in which software is conceived, tested and deployed.
In this talk I will describe my experience and the lessons learned in two multi-disciplinary projects: collaboration with zoologists for animal monitoring through sensing and with social psychologists for monitoring human interactions through mobile phones. The talk will discuss the ubiquity of requirements when mobile and sensing start being employed.

Biography Dr. Cecilia Mascolo is a Reader in Mobile Systems in the Computer Laboratory, University of Cambridge, UK. Prior to this, she was with the Department of Computer Science of University College London, UK. She holds an MSc and PhD in Computer Science from University of Bologna (Italy). Cecilia’s research concentrates on mobility data gathering, analysis, modeling and exploitation through research council and industry funded projects. Most of the projects are multi-disciplinary. Her research strategy is heavily experimental and deployment oriented. She has published extensively in the areas of mobile sensor networks, mobile network routing, realistic mobility models and social network analysis. Cecilia has served as in the Organization Committees of many mobile and sensor systems, middleware, software engineering and data mining conferences and workshops. She is on the editorial board of IEEE Internet Computing. More details of her profile are available at www.cl.cam.ac.uk/~cm542.

Proving the Shalls in Practice

Michael Whalen, University of Minnesota
Michael Whalen

Abstract: Recent advances in symbolic verification tools make it possible to rigorously analyze, and in many cases prove, whether software meets its requirements. These tools have the potential to revolutionize software verification and validation – but only if we have good requirements. Without a reasonably complete set of high-quality requirements, the benefits offered by techniques such as automated test case generation, symbolic evaluation, and proof tools is marginal.
In addition, combining symbolic verification with model-based development allows a powerful new approach for early validation of requirements. These approaches are naturally complementary: executable models often suffer from implementation bias, while declarative requirements do not. On the other hand, executable models (a la SCADE) are by construction complete and consistent, whereas declarative requirements are often incomplete and inconsistent. Using the two approaches together corrects the natural deficiencies of each approach.
In this talk, I discuss experiences at Rockwell Collins applying model checking techniques on a variety of industrial avionics models, including displays, flight guidance mode logic, and sensor fusion and voting, and the lessons learned in moving academic theory into industrial practice.

Biography Dr. Michael Whalen is the Program Director at the University of Minnesota Software Engineering Center. Dr. Whalen is interested in formal analysis, compilers, testing, and requirements engineering. He has developed simulation, translation, testing, and formal analysis tools for Model-Based Development languages including Simulink, Stateflow, SCADE, and RSML-e, and has published more than 30 papers on these topics. He has led successful formal verification projects on large industrial avionics models, including displays (Rockwell-Collins ADGS-2100 Window Manager), redundancy management and control allocation (AFRL CerTA FCS program) and autoland (AFRL CerTA CPD program). His PhD dissertation involved using higher-order abstract syntax as a basis for a provably-correct code generation tool from the RSML-e specification language into a subset of C.

Feature-Oriented Requirements: the Good, the Bad, and the Ugly

Joanne Atlee, University of Waterloo
Joanne Atlee

Abstract: A software system is often thought of in terms of its constituent features. In requirements engineering, features can serve as a shared vocabulary among stakeholders of varying backgrounds (e.g., users, developers, marketers). In design, they can form the basis of system decomposition, in which feature modules are treated as separate components that are developed in relative isolation or are supplied by third-party vendors.
The challenge of feature modularity is in managing feature interactions. Seemingly independent features may interact with each other in subtle and surprising ways. For example, a new feature may override existing behaviour, violate invariant properties, or refine the definitions of terms. Determining how interacting features ought to behave is a requirements-engineering problem. However, the scale of the feature interaction problem is non-linear in the number of features – to the point where identifying and resolving interactions dominate the feature-development process.
This presentation will give an overview of the research on feature modularity and interactions from the perspective of the feature-interaction community. It will look at general strategies for detecting and resolving classes of interactions. It will also look at some of the wicked open problems.

Biography Joanne Atlee is an Associate Professor in the David R. Cheriton School of Computer Science at the University of Waterloo. Her research interests include software modelling, automated analysis of software models, modular software development, feature interactions, and software-engineering education. She spent 10 years investigating the feature interaction problem in the telephony domain, working with companies such as Nortel Networks, AT&T Labs, and Mitel Networks. More recently, she has started to explore the modelling, analysis, and coordination of automotive features. She was Program Co-Chair for the 31st International Conference on Software Engineering (ICSE’09) and was Program Chair for the 13th IEEE Requirements Engineering Conference (RE'05). She serves on the ACM SIGSOFT Executive Committee as an at-large member and is a member of the International Federation for Information Processing (IFIP) Working Group 2.9 on Software Requirements Engineering. She is a co-author with Shari Lawrence Pfleeger on their textbook “Software Engineering: Theory and Practice”.