Integrating information sources for visualizing Java programs

A GXL Schema for Story Diagrams

“…The relationship between these source-oriented and graph-oriented representations is interesting, since the user may want to switch views between the source code and the story diagram. For example, systems like SHriMP can support a similar capability [8]. For our purpose, we need some consistent color scheme to relate elements visually in the source code view to the story diagram view.…”

Advanced widgets for Eclipse

“…Most GUI libraries contain components for Lists, Trees, Tables, Rich Text Fields, Buttons, Sliders, etc, but do not provide components for 2D/3D Graphs or UML Diagrams. If the information is complex and needs novel representations, usually the developer has to implement it themselves using 2D and 3D drawing packages (such as Draw2D [1], Piccolo [3], and OpenGL), borrowing concepts from industrial and academic visualization environments such as Tom Sawyer, Snap Together Visualization [8] and SHriMP [7]. This often produces very good tools for very speci c domains such as Software Visualization, or very generic visualization environments which fail to be practical for common adoption.…”

Software Visualisation Techniques Adapted and Extended for Asynchronous Hardware Design

“…The best reference about collaborating views is Shrimp [7], showing some real similarities with this thesis's work. Four sources of information are collected and represented in multiple views, some of them being network graphs representing the structure of the system.…”

“…Hierarchical clustering can be used to induce a hierarchy in a graph structure that might not otherwise have a hierarchical structure. Michaud et al have described Shrimp [7], which gathers various sources of information together (source code artifacts and relationships, architectural abstractions, documentation and history information, metrics and analysis information) and visualise them together. The difference between their work and the one presented here is that they limit their study to static information, while dynamic simulation data is also used here to deduce some clusters.…”