Layered Design: Concepts, Case Studies and Processes - Theories and Implementations
Many modern applications are focused on providing as many functions as possible and attract as many users groups as possible. The interfaces become very advanced but also quite complex. In many cases the design is perceived rather difficult by the end users. The challenge we and others try to meet is how to make the graphical interfaces easier to use and better adapted to the many different users and user groups.
There are a number of solutions but we have focused the concept of layered design, which is only one of many possible alternatives. The focus of a layered design is to divide the contents of the graphical user interface into parts or so called layers. The layers are arranged or ordered in a structure, where each layer contains a specific set of functions. The order is decided by an ordering parameter, which could be for example frequency of use, complexity, or level of difficulty. The arrangement of the layers and their contents should together form a sequence that provides the user with a meaningful order in which the functions should be used. Already in 1976 and 1977, the thought of users learning complex tasks in small steps was awakened and that thought was implemented within layered design. The layered design has later been seen in applications like games, web-systems, language applications or tutoring systems. A branch of the layered design concept is focused mainly on the users’ experience, learning and knowledge levels. This branch was presented in 1998 and later named Multi-Layered Design.
Focus of this thesis work is to define the layered design and the branch of Multi-Layered Design further and to practically show how it can be used through different empirical studies involving end users. From the empirical data a number of problems could be identified. The difficulty of identifying layers and their contents was the first problem encountered. To adapt layered structures to advanced situations with several user groups or complex relationships between functions was the next problem found. A third problem was how to handle layered structures when there are a large number of users with different domain knowledge and varying tasks to perform and knowledge and experience not is suitable as the main focus of the layered structure.
As a first solution a more strict process was suggested. It could handle large amounts of data, treat it in an algorithmic manner and produce guidelines for how to define the layers, their contents and in which structure they should be arranged. A second solution was a more advanced layer structure, an extended structure with parallel tracks for the different groups and a meta level of components of functions. As a third solution, an augmented version of extended layer structures was created and a layered design where several parameters can be used to order the functions by. New empirical studies were conducted to test the new solutions and an automated tool for the process was developed. To complete the presentation we will initially present the theory of the layered concept, then how it can be applied in practical cases, moving on to Multi-Layered Design theory and its practical implementations.