Paul Vickers

A survey of two-dimensional graph layout techniques for information visualisation:

Many algorithms for graph layout have been devised over the last 30 years spanning both the graph drawing and information visualisation communities. This article first reviews the advances made in the field of graph drawing that have then often been applied by the information visualisation community. There then follows a discussion of a range of techniques developed specifically for graph visualisations. Graph drawing algorithms are categorised into the following approaches: force-directed layouts, the use of dimension reduction in graph layout and computational improvements including multi-level techniques. Methods developed specifically for graph visualisation often make use of node-attributes and are categorised based on whether the attributes are used to introduce constraints to the layout, provide a clustered view or define an explicit representation in two-dimensional space. The similarities and distinctions between these techniques are examined and the aim is to provide a detailed assessment of currently available graph layout techniques, specifically how they can be used by visualisation practitioners, and to motivate further research in the area.

When Bugs Sing

In The Songs of Insects, Pierce (1949) described the striped ground cricket, nemobius fasciatus- fasciatus, which chirps at a rate proportional to ambient air temperature. Twenty chirps-per-second tell us it is 31.4°C; sixteen chirps and its 27°C. This is a natural example of an auditory display, a mechanism for communicating data with sound. By applying auditory display techniques to computer programming we have attempted to give the bugs that live in software programs their own songs. We have developed the CAITLIN musical program auralisation system (Vickers & Alty, 2002b) to allow structured musical mappings to be made of the constructs in Pascal programs. Initial experimental evaluation (Vickers & Alty, 2002a, 2002b) showed that subjects could interpret the musical motifs used to represent the various Pascal language constructs. In this paper we describe how the CAITLIN system was used to study the effects of musical program auralisation on debugging tasks performed by novice Pascal programmers. The results of the experiment indicate that a formal musical framework can act as a medium for communicating information about program behaviour, and that the information communicated could be used to assist with the task of locating bugs in faulty programs.

Musical program auralization: Empirical studies

Program auralization aims to communicate information about program state, data, and behavior using audio. We have argued that music offers many advantages as a communication medium [Alty 1995]. The CAITLIN system [Alty and Vickers 1997; Vickers 1999; Vickers and Alty 1996, 1998] was constructed to provide auralizations within a formal structured musical framework. Pilot studies [Alty and Vickers 1997; Vickers 1999] showed that programmers could infer program structure from auralizations alone. A study was conducted using 22 novice programmers to assess (i) whether novices could understand the musical auralizations and (ii) whether the musical experience and knowledge of subjects affected their performance. The results show that novices could interpret the auralizations (with accuracy varying across different levels of abstraction) and that musical knowledge had no significant effect on performance. A second experiment was conducted with another 22 novice programmers to study the effects of musical program auralization on debugging tasks. The experiment aimed to determine whether auralizations would lead to higher bug detection rates. The results indicate that, in certain circumstances, musical auralizations can be used to help locate bugs in programs and that musical skill does not affect the ability to make use of the auralizations. In addition, the experiment showed that subjective workload increased when the musical auralizations were used.

Using music to communicate computing information

Abstract The audio channel remains little used in most computing applications, often its use being relegated to providing trivial sound effects whose novelty value soon wears off. Nevertheless, in principle sound offers much to the process of human–computer interaction as for most people the notion of auditory imagery is easily accepted. In this paper we explore how sound, specifically musical sound, can be used to communicate computing information. The findings of two studies are presented. The first investigated how pitch intervals and musical phrases of complex (non-sinusoidal) tones can be recognised. The second study aimed to demonstrate that musical structures could communicate information about high-level programming language structures and program run-time behaviour. Both studies showed that music could successfully be used as a communication medium and that listeners did not need to be musically trained to benefit from the audio signals. Finally, recommendations for further work are made.

Siren songs and swan songs debugging with music

Program execution behavior can be mapped to a structured musical framework that helps locate and diagnose software errors.

Musical Program Auralisation: A Structured Approach to Motif Design

Abstract In an earlier paper, Vickers and Alty (2002) showed that musically untrained users could make use of musical cues to understand computing information. Using a technique known as musical program auralisation, they showed that music could communicate run-time and structural information about Pascal programs. This paper describes how a set of hierarchically related auralisation motifs was designed and constructed within a formal musical framework. These auralisations were then evaluated in an experiment to determine how well they could be interpreted by computer science students. The results showed that the musical motifs were generally understood by the subjects and that any prior musical training of the subjects did not affect their ability to interpret the musical signals. Based on the results of the experiment and study of some cognitive aspects of music perception, a set of organising principles for musical program auralisation is proposed. Finally, recommendations for further study are made with particular regard to assessing the usefulness of the auralisations in program debugging situations.

Using music as a communication medium

Music is a rich communication medium, and there are some similarities between the job of a music composer and that of an HCI designer (although their objectives may be different). Whilst sound has been used in interfaces, its use has mainly been at a primitive level, often involving real-world sound. Since music offers a highly structured set of mechanisms for communicating, it is surprising that there have been so few attempts at exploring its possibilities. Our current activity involves investigations into the use of music in algorithmic audiolisation and program debugging.

Parsed use case descriptions as a basis for object-oriented class model generation

Abstract: Object-oriented analysis and design has become a major approach in the design of software systems. Recent developments in CASE tools provide help in documenting the analysis and design stages and in detecting incompleteness and inconsistency in analysis. However, these tools do not contribute to the initial and difficult stage of the analysis process of identifying the objects/classes, attributes and relationships used to model the problem domain. This paper presents a tool, Class-Gen, which can partially automate the identification of objects/classes from natural language requirement specifications for object identification. Use case descriptions (UCDs) provide the input to Class-Gen which parses and analyzes the text written in English. A parsed use case description (PUCD) is generated which is then used as the basis for the construction of an initial UML class model representing object classes and relationships identified in the requirements. PUCDs enable the extraction of nouns, verbs, adjectives and adverbs from traditional UCDs for the identification process. Finally Class-Gen allows the initial class model to be refined manually. Class-Gen has been evaluated against a collection of unseen requirements. The results of the evaluation are encouraging as they demonstrate the potential for such tools to assist with the software development process.

Understanding Visualization: A Formal Approach Using Category Theory and Semiotics

This paper combines the vocabulary of semiotics and category theory to provide a formal analysis of visualization. It shows how familiar processes of visualization fit the semiotic frameworks of both Saussure and Peirce, and extends these structures using the tools of category theory to provide a general framework for understanding visualization in practice, including: Relationships between systems, data collected from those systems, renderings of those data in the form of representations, the reading of those representations to create visualizations, and the use of those visualizations to create knowledge and understanding of the system under inspection. The resulting framework is validated by demonstrating how familiar information visualization concepts (such as literalness, sensitivity, redundancy, ambiguity, generalizability, and chart junk) arise naturally from it and can be defined formally and precisely. This paper generalizes previous work on the formal characterization of visualization by, inter alia, Ziemkiewicz and Kosara and allows us to formally distinguish properties of the visualization process that previous work does not.

Pen-Based Gestures: An Approach to Reducing Screen Clutter in Mobile Computing

Mobile computing is an area of high growth despite having some serious design issues. It is difficult to increase the size of the screen because of the device’s physical constraints. Consequently, as mobile applications have incorporated more functionality, screen clutter has increased. One method of reducing clutter is to remove visual controls and use pen-based gestures instead. We describe a cinema listing application for a Palm OS device that implements pen-based gestures as the main input method. Two methods are used to communicate the options available on each screen: audio cues and small visual prompts. Preliminary results suggest that buttons can be removed from the screen without detriment to task accuracy or user performance.