The problem of disorientation
Hypermedia gives us instant access to vast amounts of information. Information items consisting of text, pictures and sound, are linked together in a network of arbitrary structure and size. These items, the nodes of the network, can be acces-sed by the user in a rapid succession. Links to related subjects can be traversed instantly. Within moments, it is possible to leap between 18th century gardening and the beginnings of life.
With the potentially large amounts of information available, and the possibility of moving rapidly through the hypermedia network, one problem becomes appar-ent: How do we prevent the user from getting lost? How should the user be able to orient herself while traversing the network of nodes and links? The problem of getting lost is known as the problem of disorientation (Conklin 1986). Conklin identifies two major problems: knowing where you are in the hyper-media network, and knowing how to find some other node in the network (and knowing if such a node exists).
A basic reason for disorientation problems is that it is difficult to remember how the potentially complex hypermedia network is structured. It can also be hard to keep track of where one is in relation to the overall structure. Espe-cially when nodes are presented one at a time only, as in for instance HyperCard (Apple 1987), it can be difficult to remember how nodes are related and to which context a node belongs. Presenting one information item only, at a given time, can make it hard to relate items to each other, and to grasp the overall structure. Furthermore, the abstract nature of hypermedia makes it potentially more diffi-cult to comprehend than, for instance, printed media. It is difficult to directly perceive the physical volume of a hypermedia system in the same way as with a book.
Getting lost can be unproductive, even fatal, in many situations. In work situa-tions which require quick decisions and instant action, like process control, it is an obvious risk that a disoriented operator can make serious mistakes since she does not know exactly what is going on. In learning situations, loosing track of where one is can result in frustration and negative learning effects. Confusion about how a subject area is structured and misassumptions about the relation between different topics could arise as a result of being disoriented. As a con-sequence the student might not only be confused, but might assimilate spurious knowledge, due to misunderstandings and misconcep-tions.
A central problem of hypermedia is that it might be impossible and/or intractable to know that you have seen everything of interest and not missed anything. After reading through a book, you can be relatively sure that you have read the whole material, or at the very least have an impression of how much material have been skipped. In a hypermedia system it is more difficult to be certain that you have found all relevant information.
Finally, negative attitudes and dislike might arise as an additional consequence of disorientation. Getting confused repeatedly can cause negative feelings and un-necessary stress. This is important both in work situations, in learning environ-ments, and when using information systems at museums and other public loca-tions. If the user feels uncomfortable and confused, she is likely to loose interest and discard the system.
Orientation and visual communication
In a hypermedia system, the interaction between the user and the system is based mainly on visual communication:
Figure 1.1: Good visual communication can improve orientation. Visualized spatial information has a potential for orienting the user that goes beyond alphanumeric messages. If information is visualized the user does not have to remember where in the hypermedia structure she is - it is shown right before her eyes. Future hypermedia systems will develop a visual language, designed to solve disorientation problems unique to hypermedia, that goes beyond that of, for example, printed media and film. (Illustration by Gösta Lindwall, Copyright © 1990)
The problem of disorientation seems to arise more frequently in hypermedia than in, for instance, printed media. This is in part related to basic differences betwe-en the two media forms; hypermedia is more abstract and difficult to grasp than printed media. An additional reason is that printed media has a long tradition in communication. The author of a book knows that if she does not communicate well, the read-er will feel confused and loose interest.
While graphical user interfaces have received a dramatic increase in interest during the recent years, there is still a substantial difference in sophistication between the visual communication used in even 'state of the art' user interfaces, and that of illustrated encyclopedias, magazines, and animated cartoons.
We might be able to reduce some disorientation problems of hypermedia by learning from how information is communicated in traditional media. Two important principles of communication is what to communicate and how to communicate it (which in turn depends on who we want to communicate with). Thus, we must investigate what information is needed for the user to orient herself. We will call this information orientational information. The user might for instance get lost if the present location and the structure of the hypermedia network is not known. In chapter 3 we will examine this issue in detail.
Furthermore, we must decide how to communicate orientational information to the user. In this thesis we will examine a technique for presenting information called lexivisual presentation (Lidman & Lund 1972). Lexivisual presentation is a set of principles and techniques for how to present information (with text and pictures) in printed media. Central principles are: focus on the primary message, and present detailed information in the context of the whole. This can reduce the risk of loosing track of what one is reading. Often graphical techniques like zoom-boxes, miniaturized maps, and speech balloons are used. Lexivisual presentation is used in, for instance, modern encyclopedias, comic books and instruction manuals.
How communica-tion principles of traditional media can be applied to computerized media has been studied previously by, among others, Heckel (1984), Wise & Debons (1987), Kindborg & Kollerbaur (1987a, 1987b), Ferm et alia (1987 1988), Kollerbaur, Larnhed, Kindborg (1989), Nielsen (1989), Kindborg (1990), and Kollerbaur & Kindborg (1990).
Purpose of the thesis
The purpose of this thesis is to study if and how principles and visual techniques for orientation used in printed media can be applied to hypermedia. In particular, lexivisual presentation techniques and principles are studied. Orientation in hypermedia systems is also analyzed.
Restrictions
This thesis is restricted to reading of hypermedia, that is browsing and naviga-tion, not authoring. Thus, problems which arise in situations like writing, plan-ning, idea processing and similar tasks, are not studied. Although disorientation problems certainly occur in authoring situations, this study is is focused on browsing and navigation. However, the basic principles discussed in this report are likely to be applicable to authoring as well.
Moreover, disorientation problems related to the literal content of information contained in a hypermedia system are not explicitly dealt with. Focus is on how to present infor-mation to ease orientation in hypermedia systems. That is, focus is on the form of the presentation of orientational information, rather than on the literal content. This is important since a poor presentation can make the subject very hard to understand. It should be emphasized, however, that content and form are very closely related in practice. The structure of the subject area has a significant impact on the design of the system. In fact, the subject area will to a large extent determine how information is communicated.
Finally, I have chosen to focus on visual communication principles employed by printed media. While many other forms of communication are of interest, like film and drama, a detailed discussion of them are beyond the scope of this thesis.
Research methods
Several different methods have been employed when conducting this research. A number of hypermedia systems and tools have been studied and analyzed. Examples of systems studied are NoteCards (Halasz et alia 1987), Guide (OWL 1988), HyperCard (Apple 1987), SuperCard (Silicon Beach 1989), KMS (Akscyn et alia 1988), Plus (Format 1990), ToolBook (Asymetrix 1990), and numerous applications created using these systems. In parallel, visual presen-tation techniques and their application to hypermedia have been studied. This work has been going on between 1987 and 1990 and forms the basis for this thesis.
Furthermore, various design examples have been discussed and analyzed in cooperation with hypermedia researchers, media experts and graphical design-ers. An empirical study has been conducted to test the assumptions that it is of importance to orientation to give an overall view of the hypermedia network and to show detailed information in its context. The experimental set-up used in this study and the results from it are described in chapter 6.
Overview of the thesis
The subject of this thesis is how principles for orientation used in printed media can be applied to hypermedia. Chapter 3 analyzes the problem of disorientation, and chapter 4 presents principles for orienting the reader of printed media. These two chapters form a basis for subsequent chapters. In chapter 6 a case study is presented. In Chapter 7 possible enhancements of hypermedia structures are discussed. An object-oriented data model for hypermedia, which is better suited for the kind of presentations proposed than the traditional node-link model, is also suggested.
The contents of each chapter is summarized below.
Chapter 1, Introduction, gives a background to the study and briefly presents the problem of disorientation.
Chapter 2, Hypermedia - a summary of concepts, gives a summary of the basic concepts of hypermedia.
Chapter 3, The problem of disorientation, provides an analysis of the problem of disorientation. Potential causes for disorientation are identified.
Chapter 4, Communication with text and pictures, goes into detail of how orientation is given in printed media like books, comics, newspapers, et cetera.
Chapter 5, Present techniques for orientation, analyzes orientation techniques commonly used in hypermedia systems.
Chapter 6, The solar system - an explorative study, presents a comparative study of two hypermedia systems. One system is designed using lexivisual principles, the other one is not.
Chapter 7, Enhanced hypermedia structures, discusses how lexivisual principles could be used to further enhance orientation in hypermedia structures. A new object-oriented data model for hypermedia tools, making it easier to implement the designs discussed, is also presented.
Chapter 8, Summary, summarizes the results of this thesis. Ideas for future research are also presented.
Definitions, at the end of the thesis, provides a summary of important terms and their interpretation.