User interfaces are getting more and more complex, and adaptable and adaptive interfaces have been proposed to address this issue. Previous studies have shown that users prefer interfaces that they can simply adapt themselves to self-adjusting ones. However, most existing user interface toolkits provide very little support for creating adaptable interfaces. As a consequence, interface customization techniques are still very primitive and usually constricted to particular applications. In this paper, we present User Interface Façades, a system that provides end-users with simple ways to adapt and recombine existing graphical interfaces, through the use of drag-and-drop paradigm. User Interface Façades employs a more appropriate level of granularity for adaptation compared to previous work and also allows end-users to adapt the interaction of arbitrary applications. Finally, we show several examples to demonstrate the power of the new technique. ACM Classification: H5.2 [Information interfaces and

When involved in the visual design of graphical user interfaces, graphic designers can do more than providing static graphics for programmers to incorporate into applications. We describe a technique that allows them to provide examples of graphical objects at various key sizes using their usual drawing tool, then let the system interpolate their resizing behavior. We relate this technique to current practices of graphic designers, provide examples of its use and describe the underlying inference algorithm. We show how the mathematical properties of the algorithm allows the system to be predictable and explain how it can be combined with more traditional layout mechanisms. ACM Classification: H5.2 [Information Interfaces and

SVG—a W3C XML standard—is a relatively new language for describing low-level vector drawings. Due to its cross-platform capabilities and support for events, SVG may potentially be used in interactive GUIs/graphical front-ends. However, a complete and full-featured widget set for SVG does not exist at the time of this writing. I have researched and implemented a framework which retargets a complete and mature rasterbased widget library—the JFC Swing GUI library—into a vector-based display substrate: SVG. My framework provides SVG with a full-featured widget set, as well as augmenting Swing’s platform coverage. Furthermore, by using bytecode instrumentation techniques, my Swing to SVG bridging framework is transparent to the developers—

Abstract. Graphical rendering must be fast enough so as to avoid hindering the user perception/action loop. Traditionally, programmers interleave descriptions and optimizations to achieve such performances, thus compromising modularity. In this paper, we consider graphic rendering as a compilation chain: we designed a static and dynamic graphical compiler that enables a designer to clearly separate the description of an interactive scene from its implementation and optimization. In order to express dependencies during run-time, the compiler builds a dataflow that can handle user input and data. We successfully used this approach on both a WIMP application and on a demanding one in terms of computing power: description is completely separated from implementation and optimizations while performances are comparable to manually optimized applications.

Abstract—The layout mechanisms for many GUI toolkits are hard to understand, the associated tools and API’s often difficult to use. This work investigates new, easy-to-understand layout mechanisms and evaluates its implementation. We will analyze the requirements for the definition of layouts of a graphical user interface. Part of the issue is that several aspects need to be considered simultaneously while laying-out a component: the alignment with other components as well as its own behaviour while resizing its container. Moreover, the used tools should isolate the designer/drawer from the implementation details of the framework. We present the details of our new GUI layout system, discuss the choices we made for our new layout algorithm and detail implementation issues. Moreover, we present also the user interface for our new GUI builder system that contains several innovations, such as a preview window to show the effects of layout configuration choices in real-time. We present an evaluation of our new system by attacking the complex GUI layout problem mentioned above. I.

In this paper we argue that concrete User Interfaces (UI) languages are not suitable for adaptation. In addition, we point out the fact that the quality of tailored UIs is far better than the quality of automatically generated UIs. Therefore we propose to capitalize human designed UIs in a structured knowledge base as promoted by Service Oriented Approaches. The base aims at supporting designers and/or automatic UI generation algorithms in retrieving UI descriptions both at design time and runtime.

Abstract. Building interactive software is a notoriously complex task, for which many programming tools have been proposed over the years. Although the research community has sporadically identified usability requirements for such tools, tool proponents rarely document their design processes and there is no established reference for comparing tools with requirements. Furthermore, the design of most tools is strongly influenced by the design of their underlying general purpose programming languages. These in turn were designed from their own set of little-documented requirements, which adds to the confusion. In this paper, we provide a review and classification of the requirements and properties expected of interactive development tools. We review how designers of APIs and toolkits for interaction-oriented systems set the usability requirements for the programming interface of their systems. We relate our analysis to other studies in related domains such as end-user programming, natural programming, and teaching. 1