TiChi Introduction

GOMS
Fitt's Law
Object-Action Interface
Prescriptive Theories
Fisheye strategy
Conceptual, semantic,
    syntactic, & lexical

Direct Manipulation

Information Processing
Hacker's Action Theory
Attention & Memory
Andersen's ACT-R
Knowledge & Mental
Social & Cultural

Theories in Computer human interaction
Printer Friendly Version
Fisheye Strategy
Cassie Thomas
cassie@cs.umd.edu
October 28 2002

Overview

Often when a person is describing an object or a location to another, they describe an overview of the object before getting into the details. Essentially a local detail and a global context are presented. Our knowledge of things around us, constantly changes, and no matter where we are some things will be more important to us than others. Concepts that are similar in nature, or directly related to each other help us to remember them easier.

This technique of leaving out detail when dealing with things in large scale is seen in how people talk, in art, and in user interfaces. Furnas' uses this concept to develop the fisheye strategy.

Two points of interest are noted when dealing with this strategy: 1) Accessing large structures can take a lot time, and
2) Problems can be solved by leaving out detail the farther away from the
focus point.

The key concept of the fisheye strategy is that objects close to the focal point are clear and distinguishable to the user, and as objects move further away from the point of interest, the level of detail decreases the farther away you get from the focal point. The fisheye strategy creates an overview, while maintaining details at the same time. The area of interest is magnified to show the detail, while the context is maintained, so continuity is preserved at the periphery.

Scope, Application, and Limitations

The fisheye strategy presents a way to present a small display of a large structure. These structures can be network trees, programs, entire systems, and databases. While the amount of information being presented to the user increases, the viewing space (i.e. computer screen) remains relatively small. This can present an interface design problem, for developers of systems. The idea is to present a large amount of data to users in a way that is searchable, and getting information is not too timely of a task.

Furnas' fisheye strategy attempts to solve this problem by providing a local context against a global context. This is a focus and context visual technique which can often be referred to as an 'attention-warped display'. Two resources are considered here: the users attention, and the screen space. This strategy considers the screen space so that more of the resource is available for the details most important the user.


Figure 1: Map of Washington D.C. [5]

There are, however, some limitations of this strategy. The fisheye strategy can be appropriate in a system if distortion is not an issue for objects distant from the focal point. The effects of distortion can be hard to predict and control. Distortion can disorient the user, and the information at the farthest distances can be lost. Also, the notion of distance is available, but it is not so clear about what determines the level of detail.

Principles

The fisheye strategy identifies three properties that must be defined:
1. Focal point
2. Distance from focus: D (.,x)[D(.,.)=0]
3. Level of detail, importance, resolution: LOD (x)

A point of interest has to be defined so that the interaction with the focus, meaning what is going to be the global context, can be determined. The 'distance from focus' concept determines the distance from my point of interest (focus) to some point x. Examples of 'Distance from focus' could be the distance from the McDonalds in a neighborhood (the McDonalds being the point of interest) to a particular house, or from a users home directory to four levels deep in the file structure.

Degree of Interest is another concept in fisheyes. The idea is that a user at any given point of interaction within a system, is not going to be interested in the entire system all of the time. For a particular purpose, it is necessary to determine how interested a user is in an application on the system. The strategy would display applications in the system that are of most interest to the user in great detail, while applications not used often would be in less detail.

A higher degree of interest is indicated by a higher value. The degree of interest at a line, x, given, the current point, is defined to be [1],
DOI (x|.)=F (LOD (x), D (.,x))

The DOI is composed of a static component and the dynamic component. The static component is the priori importance, or the global importance of the element relative to every other object in the system. For the user, the global importance is how an object is used more than another object in the system. The dynamic component creates a relationship between the users interest and the importance of an item depending on the latest interactions on a system. The DOI is assigned to every element in the users information space, and a node is selected as the central focus point. It is important to notice that if the point of interest changes, then the DOI must be recalculated for every node [5].

Example

An example of Furnas' fisheye strategy is the fisheye menu, which applies the fisheye graphical visualization techniques to linear menus, thereby allowing for efficient searching through long menus. The size of the menus are changed, so depending on what the users DOI is the menu will change dynamically to provide a point of interest around the mouse pointer. This strategy makes it possible to present the whole menu to the user without having to use a hierarchy, or scrolling, to lose the users attention and focus. [3]


Figure 2: Fisheye Menu [6]

The fisheye strategy has also been applied to abstract graphs, structured data trees, geographical views, and maps. Furnas' used the fisheye strategy to make searching through a hierarchical structure in source code easier. In this case, Furnas used line numbers to calculate the distance from the point of interest to every node.

Semantic Fisheye Views (SFEVs) and Graphical Fisheye Views (GFEVs) are other examples of fisheye views. The SFEV is a context-aware tool to search, browse, organize, and synthesize large collection of multimedia documents. It focuses on annotated images in particular. The GFEV is an analogy to the fisheye lens. It distorts the characteristics of an image; for example a city map. The graphical fisheye view, will concentrate on an object or an area as the point of interest, and depending on the graphical distance, the distortion will be determined.

Applicability to HCI

This strategy has influenced many areas of research within HCI. The concept of fisheye menus, in system applications has increasingly become an area of interest. As well as, displaying network topologies, and large structures to users in such a way that is most effective is presenting the information. The focus+context concept was spawned under the fisheye concept and has led to extensive research in information visualization. When interfaces and visualizations follow human perception, intuition about the data presented to the user is optimal. As humans, we perceive information in a context+detail orientation. Objects at the center of our retina are at a high resolution, while objects at our periphery are seen in far less detail [5]. Studies have shown that the fisheye strategy increases user performance on system tasks. Tasks were completed in fewer steps with less navigational error. In systems where user performance is critical, designers might find it ideal to implement interfaces with the fisheye strategy in mind.

References

[1] Furnas, George, 1986. Generalized Fisheye Views. Human Factors in computing systems, CHI '86 conference proceedings, ACM, New York, pp. 16-23.

[2] Shneiderman, Ben. 1998. Designing the User Interface: Strategies for effective Human-Computer Interaction. Addison-Wesley, Reading, MA.

[3] Card, S., Mackinlay, J., and Shneiderman, B.1999. Readings in Information Visualization: Using Vision to Think. Morgan Kauffman Publishers, San Francisco.

[4] Furnas, George, 1981. The Fisheye View: new look at structured files. Bell Laboratories technical memorandum, #81-11221-9.

[5] http://web.mit.edu/16.399/www/course_notes/context_and_detail1.pdf. October 23, 2002

[6] http://www.cs.umd.edu/hcil/fisheyemenu/ October 23, 2002.