Abstract

User-Centered Design (UCD) is a philosophy surrounding interactive system design, with the purpose of achieving product usability. One challenge with UCD and its related methods is the lack of a concrete process which supports designers in building user interface (UI) designs founded on user experiences. In current practice, design decisions are made based on loosely-defined guidelines, giving rise to a significant "gap" between user analysis and design outcomes. This is especially problematic for novice designers who lack the background and training required to make trade-offs, judgments and interpretations towards a usable design. In this thesis, we propose a Pattern-Oriented UI Design method which is driven by user experiences. It is founded on a set of core UCD principles which we have enriched with "engineering-like" concepts such as reuse and traceability. The method is based on two key artifacts--personas, used to model user experiences, and patterns, used to capture best design practices. Following this method, we define the UX-P Process, a systematic process which is semi-automated and characterized by rigorously-defined steps; designers iteratively create personas, select patterns, and compose patterns into a comprehensive design, based on user specifications and usability considerations. We have built a supporting tool, which allows designers to cluster users into personas and select candidate patterns based on persona specifications. We carried out two empirical studies with end-users. The goal of the first study was to assess the feasibility of the method; the second, to validate the process. Both studies were carried out with Bioinformatics applications and were comparative in nature testing the original design with our prototype. The outcome of these empirical studies indicated a positive increase in usability measures for our design prototypes, including a significant improvement in task times and user satisfaction.