Sebastian Feuerstack @ Fundação Armando Alvares Penteado

Developing an interactive application that should be run on several devices and even should support access via several modalities can be a tedious and cumbersome task. The heterogeneous landscape of devices available, equipped with different operating systems, requiring applications to be implemented with different computer languages, different user interface toolkits, need to follow different interaction patterns and style guides, usually requires implementing the same software more than once and raises the costs of maintaining these different realizations of the same applications afterwards as well.

Model-driven design and development of user interfaces (MDDUI), which is practiced since the late 1980’s in Human-Computer Interaction, seems to be a promising approach to support software developers, developing interactive applications. MDDUI aims at modelling the different aspects of the user interface on certain levels of abstraction and offers a declarative way of modelling multi-platform user-interfaces. Each model is refined to more specific ones. By this approach the more abstract models can be shared for the development of several platforms and modalities. The more concrete ones are used to address certain capabilities of specific platforms. MDDUI is concerned with two basic problems:

1. The identification of suitable sets of model abstractions and their relations for analyzing, designing and evaluating interactive systems.
2. The specification of software engineering processes that change the focus from (manual) implementation to the tool-driven design of models.

One often mentioned disadvantage of MDDUI approaches is the long, often very abstract and complex process requiring the developers to learn new languages and concepts (such as task and dialogue models, abstract and concrete user interfaces) and to follow time consuming processes and tool chains until first results can be inspected. These are often not the expected ones since even for experienced developers it is often hard to mentally bridge the gap between the manipulations on a high level of abstraction and the concrete outcome for a whole bunch of different devices and modalities.

My research interests about MDDUI are focused on developing resolutions to bridge the gap between the abstraction of models and concrete realizations of user interfaces as well as closing the technical gap between what’s being designed and what’s realized by an implementation. Therefore I am researching about methods that offer intermediary steps allowing deriving prototypes as early and often as possible. Further on, I am interested in realizing runtime environments that can directly execute the design models and in collaborative design environments that are targeted to support interaction designers explicitly and additionally to software developers.

2008

• Sebastian Feuerstack; A Method for the User-centered and Model-based Development of Interactive Applications, Institut für Wirtschaftsinformatik und Quantitative Methoden, PhD Thesis, http://opus.kobv.de/tuberlin/volltexte/2009/2141/, Technische Universität Berlin.
• Sebastian Feuerstack, Marco Blumendorf, Maximilian Kern, Michael Kruppa, Michael Quade, Mathias Runge and, Sahin Albayrak; Engineering Interactive Systems 2008: 2nd Conference on Human-Centered Software Engineering (HCSE- 2008) and 7th International Workshop on Task Models and Diagrams (TAMODIA-2008). TAsk MOdels and DIAgrams (TAMODIA-08); 2008, September 25-26, Pisa, Italy, Pages 134-141, Lecture Notes in Computer Science (LNCS), Vol. 5247, ISBN 978-3-540-85991-8, Springer, Berlin, 2008
• Grzegorz Lehmann, Marco Blumendorf, Sebastian Feuerstack, and Sahin Albayrak; Utilizing Dynamic Executable Models for User Interface Development; Interactive Systems. Design, Specification, and Verification: 15th International Workshop, DSV-IS 2008 Kingston, Canada, July 16-18, 2008 Revised Papers, Pages: 306 – 309, 2008 ISBN:978-3-540-70568-0
• Sebastian Feuerstack, Marco Blumendorf, Sahin Albayrak; Bridging the Gap between Model and Design of User Interfaces; Workshop “Modellbasierte Entwicklung von Interaktionsanwendungen”, INFORMATIK 2006 Conference, Oct. 2-6, Dresden, Germany