Keynote Lectures

Domain Modeling
Stephen Mellor, Freeter, United Kingdom, United Kingdom

Data, Context, Situation - On the Usefulness of Semantic Layers for Designing Context-aware Systems
Daniela Nicklas, Carl von Ossietzky Universitaet Oldenburg, Germany, Germany

Empathy-driven Design for Emotional Wellbeing
Fabio Casati, University of Trento, Italy, Italy

Enterprise Ontology Driven Software Generation
Jan Dietz, Delft University of Technology, Netherlands, Netherlands

Abstract
Tying together many aspects of software and system engineering is the concept of a subject matter or domain. Short visits to Mr Google and Ms Wiki reveal many near synonyms: conceptual model, ontology, problem domain, concerns, semantics, etc. Each term is defined within their own context: possibly equivalent, certain to be treated separately within, er, their own domains. Yet rarely do we see models of the domains in a system. Instead, we see--overwhelmingly--models of software or other implementation mechanisms. We often see appalling mixtures of domains in system models. This makes it difficult to work independently, ties an application unnecessarily to services, and makes subject matters difficult to reuse and compose. Even less do we see explicit discussion within projects of how best to select and organize the domains in a system and the relationships between them. The keynote presentation will discuss how to model the domains in a system, the relationships between them, and how to work with a domain chart on a project.

Abstract
Context-aware applications adapt their behavior according to the current situation of their user or their (often physical) environment. This adaptation could be the change of the user interface, the performance of actions (like sending messages or triggering actuators), or the change of used resources (like network bandwidth or processing power). To determine relevant situations, many heterogeneous data sources could be used, ranging from sensor data over mined patterns in files to explicit user input. Since most sensors are not perfect, context quality has to be considered. And since many context-aware applications are mobile, the set of data sources changes during runtime. All this issues make context management and reasoning, and the development of correct adaptations within context-aware applications a challenging task. This keynote introduces a layered model to separate different tasks and concerns in designing data models for context-aware applications. It shows how existing works map to this layered model, and how the model can help in designing context aware applications that are better to maintain and safer to use.

Abstract
In this talk I will discuss how to design applications that can help people, and specifically older adults, improve their emotional wellbeing - or, in bold terms, happiness. Despite the obvious importance of the topic and the terrific impact this line of research can have for society, it is surprising how little progress has been made in the last decade, both in terms of research and in terms of products. In the talk I will first elaborate on why this is the case and exemplify the main flaws in the design thinking process that led to the current state of affairs. I will then discuss how to approach the design of applications for emotional wellbeing, stressing that the key is the need for every single designer and programmer to develop empathy with the people they are trying to help. I will then move on to discuss which are the main characteristics an application should possess, introduce some example applications and report on experiments, results, and findings in different contexts.

Abstract
Model Driven Engineering has been with us for quite some time, the most well known approach being OMG's Model Driven Architecture. However, although it has brought substantial benefits compared to other software engineering approaches, Model Driven Engineering presently still suffers from two major shortages. First, it is unable to deliver domain models from which the correct functional requirements can be derived. Hence, true validation is hardly possible: the software does not meet user expectations. Second, the models to be produced during the system development process, are not formally defined. Hence, their verification remains a cumbersome task. One of the theoretical pillars of Enterprise Engineering (EE) is the Generic System Development Process. It distinguishes between the using system and the object system (the system to be built), and it states that any software development process should start from the ontological construction model of the using system. In addition, EE’s systemic notion of Enterprise Ontology offers a formalized ontological model of an enterprise that satisfies the C4E quality criteria (coherent, consistent, comprehensive, concise, and essential). An operational application software generator will be presented that takes this ontological model, with some extensions, as source code input and executes the model as a professional software application. Changes in the software, as required by any agile enterprise, are brought about ‘on the fly’, through re-generation, based on the modified ontological model of the enterprise.