Book Web site [may not be live] and at Amazon UK

Edward Tufte: a pioneer

Envisioning Information, The Visual Display of Quantitative Information and The Visual Display of Quantitative Information

Graphics Press Web site

Pioneers in digital media

Readings in Information Visualization: Using Vision to Think Stuart Card, Jock Mackinlay and Ben Shneiderman (Morgan Kaufmann, 1999 )

A collection of papers that present the research and provide the foundation for the field of information visualisation. The editors are veterans of the discipline and a number of projects in which they have been involved have since been commercialised – a testament to their value. The extensive introduction presents a thorough overview of the field of information visualisation, but this book is really aimed at researchers, people pursuing study or practicing in this field.

Book on Amazon UK

Map metaphors for information spaces

See Maps.net

Searching large data sets using direct manipulation

A researcher at the University of Maryland HCI Lab created an interface for finding films. Film Finder presents a range slider that allows filtering by film length (a technique known as ‘direct manipulation’) and year. Buttons allow for selection by ratings and awards. Query results are immediately displayed as a ‘starfield’, which allows for the user to easily narrow down the selection further for themselves and see emerging patterns. A click on one of the spots displays a box with details of each film and a picture of one of the actors. The concept has now been commercialised as Spotfire.

Navigation with focus and context

Inxight’s tool, originally developed at Xerox PARC, presents a fish-eye view on a Web site structure, known as a ‘star tree’. This approach is referred to by creator Ramana Rao as a “focus plus context driven”.

Visual I|O: Decision making dilemmas

As part of their development of an information visualization system for a pharmaceutical research and development organization Visual I|O developed a non-pharmaceutical example of their approach that addressed a ‘stop or go’ decision with data to support an historical analysis. The challenge of deciding whether to pull a pitcher from a baseball game was chosen.

The base data for the example is drawn from game statistics. A dynamic ‘opinion gathering’ interface allows users to manually adjust weightings, using direct manipulation (akin to Spotfire).

The central element of the interface is a circular diagram which shows an aggregate ‘finding’ of whether the pitcher should be removed from the game in this circumstance. Each ‘slice’ of the circle represents the data from one of the criteria. The size of the slice represents its weighting, and the radius shows the value of that factor.

The criteria are grouped into shades of gray and there is a median represented by the light blue transparency. Yellow is used to indicate that at tranche is being edited. The more filled the circle, the stronger the suggestion that the pitcher should be pulled.

This demonstration tool, which was created in Flash, can be found at www.manifesto9.com/baseball

Brad Paley: TextArc

It is hard to imagine an English department getting excited about software application but TextArc, developed by New York-based interaction designer Brad Paley, brings a whole new facility to textual analysis. Originally developed to help people get their bearings in text documents, TextArc, which Paley describes as a ‘visual index/concordance’, can be used to analyze any text – though most examples to date have been taken from Project Gutenberg.

The entire text is arranged in an arc, stepping clockwise and starting at 12:00. Parallel lines of text are drawn around the outside with words apparently scattered around the inside, those that occur more often appearing brighter. The location of words is determined by averaging their position between their appearances in the text, such that a word which appears frequently towards the end of a text – for instance ‘socialism’ in The Communist Manifesto – is placed in the region of 9:00.

Colour is used judiciously. When a line of text from the outside is selected its constituent words are highlighted green. When an individual word is selected other forms of word, for instance those with the same root, are similarly picked out.

TextArc allows the reader to understand the subject matter and structure of a text, to get a sense of its character and pace, to begin to identify its key ideas and the style of writing. It is no substitute for reading the linear text but it provides a fascinating sideways-on look at what the seasoned reader might have thought was familiar prose.

This example and many others can be found at www.TextArc.org