Short Bio
Lorenzo Cantoni graduated in Philosophy and holds a PhD in Education and Linguistics. L. Cantoni is professor at the University of Lugano (Switzerland), School of Communication Sciences, where he teaches eLearning and eGovernment. He is also vice-director of the Institute of Institutional and Educational Communication, director of the laboratories webatelier.net: production and promotion over the Internet, NewMinE Lab: New Media in Education Lab, executive director of TEC-Lab: Technology Enhanced Communication Lab and co-director of eLab: eLearning Lab. His research interests are where communication, education and new media overlap, ranging from computer mediated communication to usability, from eLearning to eGovernment. Lorenzo Cantoni is research professor at the Center for International Health Services Research & Policy, in the Washington State University and collaborates with the Politecnico di Milano (Como campus, Italy) and the University of the Holy Cross (Rome, Italy). L. Cantoni is also President of the I.Re.F. - Lombardy Region Institute for Education and Training of Public Administration, the second largest Italian public administration school.

Abstract
Many times a design choice has a deep impact onto future website management activities, which in their turn mean allocating resources. These choices are to be verified, confirmed or refused not only against design quality and hypotheses, but mostly against actual usages. The actual usages of a website are of the utmost importance to infer and understand interests, goals and styles of users, and are to be interpreted in order to maintain, refine and enhance the website itself. It is neither a one-off nor a one-way path, but a continuous dialogue among different people and stakeholders, requiring endless hypothesizing and testing of hypotheses (e.g.: did they leave that page soon because it wasn’t relevant to them, or because they found it so relevant to print it out for further reading? Did they leave the website after accessing that page because their interest was fully satisfied, or because they didn’t find anything useful? Is a page seldom accessed because it is not that interesting or because there is a cumbersome navigation? etc.). The answers to those questions force to re-think communication strategies, as well as all other design dimensions. Moreover, an online application yields to many further exchanges, like buying, voting, subscribing, chatting, gambling, reserving, sending emails etc., activities that leave traces offering insights on our readership/users/clients, and need to be interpreted and to feed back into the website maintaining and improving processes. An overview of strategies, focuses and analysis tools will be offered, together with some fresh examples of using those data in website evolution.

Short Bio
Kostas Kontogiannis is an Associate Professor at the Department of Electrical & Computer Engineering at the University of Waterloo, Canada, where he is leading the Software Re-engineering group. Kostas received a B.Sc. degree in Mathematics from the University of Patras, Greece, a M.Sc. degree in Computer Science from Katholieke Universiteit Leuven, Belgium, and a Ph.D. degree in Computer Science from McGill University, Canada. Kostas is working in the areas of software analysis, software evolution, and software systems integration. He has been the recipient of three IBM University Partnership Awards and a Canada Foundation for Innovation (CFI) Award. Kostas is a visiting scientist at the IBM Center for Advanced Studies in IBM Toronto Laboratory, and a former member of the IEEE Distinguished Visitors Program. Kostas can be reached by e-mail at kostas(at)swen.uwaterloo.ca

Abstract
Model-driven techniques have been proposed and promoted by the Software Engineering community over the past few years as a mechanism for streamlining the design, implementation and evolution of large software applications. The basic idea behind model-driven techniques is that, design artifacts of large software applications can be represented as a collection of models which can be consequently transformed and evolved to generate specific design artifacts and even source code that complies with specific programmatic paradigms and patterns.

Even though model-driven frameworks have caught the attention of the software engineering community as a way to increase programmers’ productivity and overall system robustness through the disciplined manipulation and transformation of models and ultimately code generation, they have remained so far only in the form of “guidelines” or “standard practices”. In this respect, important questions regarding to what types of models are required for system representation, how transformations are encoded and enacted, how model constraints are denoted and validated, and how source code is generated, is left to software vendors, software architects and software developers to further design and implement.

In this keynote presentation we will focus on the challenges, issues, emerging research topics and practical examples pertaining to the use of model-driven techniques for the design, analysis and evolution of network-centric, web-based applications. Some of these challenges in such systems include the use of multi-language paradigms, the problem of maintaining consistency between various models during system evolution, dealing with underlying technology changes, and facilitating end-product customizability