Visualisation for informed decision making; from code to components

Abstract

The problem of trying to view and comprehend large amounts of data is a well-known one. A specialised variant of this problem is the visualisation of software code and components for the purposes of understanding, decision-making, reuse and even integration. In particular the visualisation of software components, at a much higher level than source code, has received very little research. Visualisation is a powerful tool in situations such as this. This paper presents the application of real world metaphor based visualisations that address this problem.The application of visualisation to selecting software components is especially novel. It seeks to decrease the effort required by system integrators when locating suitable components in what is an increasingly crowded marketplace. Accurate information and understanding are vital if correct and informed decisions and judgements are to be made.

The City Metaphor in Software Visualization

“…The work of Knight and Munro was later subsumed by a tool called Component City, which depicts components as buildings, a larger granularity shared by most other research projects that employ the city metaphor [Charters02]. Component City was not based on the Software World implementation, but instead uses XML and XSLT to generate VRML that can be viewed in browsers with a suitable plug-in.…”

Understanding software evolution with software cities

“…In ref. 8 a variant of this metaphor is presented to visualize more abstract software components in component cities.…”

An Overview of 3D Software Visualization

“…Many representations for visualizing software have been proposed. For instance, some visual representations are based on abstract shapes, such as graphs [83], trees [44,92] and geometric shapes [18,21], and others on real world objects, like 3D cities [93][94][95], solar systems [96,97], molecules [98], video games [99,100], metaballs [32], 3D landscapes [52], and social interactions [75], among others.…”

“…In addition, several systems provide semantic zoom: level of detail changes as user inspects nodes closer. To be specific, as the viewpoint moves towards a particular object, at a certain threshold, the low detailed object representation will be replaced with a high detailed representation [23,91,94]. For example, a system may display a class using a box representation at a high level of detail, and as the user gets closer it may represent additional information such as methods, variables, and software metrics.…”