Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

We need non-formal methods based on formal models in interaction design

: Hess, Steffen; Meier, Andreas; Trapp, Marcus

Kaiserslautern, 2011, VII, 4 S.
IESE-Report, 088.11/D
Reportnr.: 088.11/D
Fraunhofer IESE ()
user experience; interaction design; interaction concept; best practice; conceptual model; formal method; informal method

Motivation - With collaborative interaction design, formal models have to be used by expert stakeholders in order to work effectively and efficiently. But these formal models should not be discussed with non-expert stakeholders. Therefore, we use non-formal methods which are based on our formal models when we talk to non-expert stakeholders.
Research approach - Expert stakeholders have to get a common ground for their discussions. To achieve that common ground, expert stakeholders have to use models, which formalize an interaction they want to design. We develop non-formal methods based on a conceptual model that are easy to apply by non-experts. This way, non-expert stakeholders can allow free play to their thoughts and do not have to adopt formal entities and models. Therefore, the expert can easily interpret the information received from non-expert stakeholders by using the underlying conceptual model.
Findings/Design - Early collaboration with non-expert stakeholders in order to design interaction in a user centered way is important and performed best with a formal model for expert stakeholders and non-formal methods based on these formal models for discussions between expert and non-expert stakeholders. Actually, formal modeling is crucial from our point of view, but we experienced that insisting on it when collaborating with non-experts, leads to insufficient results.
Take away message - Without having formal models in mind, non-formal methods cannot be applied effectively for a collaborative design of interactions. Non-formal methods like an open requirements elicitation in form of workshops or interviews may fail gaining all necessary information if there is no formal model which builds the basis for the non-formal methods. Only with formal models, we are able to plan, lead, and analyze non-formal methods in a way to get optimal results.