Designing Information Spaces: The Social Navigation Approach

There are many changes happening in the world of computers and communication media. The Internet and the web, of course, have created this vast network of interlinked machines. Computers are becoming increasingly ubiquitous; they are “disappearing” into everyday objects. They are becoming increasingly small, so much so that they are now wearable. They are increasingly able to communicate with each other.

As humans, we are fundamentally social creatures. From birth we orient to other people, and as we develop we acquire abilities for interacting with one another, ranging from expression and gesture through spoken and written language. As adults we are exquisitely sensitive to the actions and interactions of those around us. Every day we make countless decisions that are shaped by our social context. Whether it is wrapping up a talk when the audience starts fidgeting, or deciding to forgo the grocery shopping because the parking lot is jammed, social information provides a basis for inferences, planning and coordinating activity.

The advent of the web has made possible the publication of vast amounts of new information. Information spaces have become more and more crowded with information. Web users are faced with the daunting task of sorting out from among all of the available information that which is most valuable to them. Social navigation systems have emerged as a broad array of techniques that enable people to work together to help each other find their way through the crowded spaces. In a social navigation system, each user who visits a web site does a small amount of work to untangle which of the paths from that web site are most valuable. Early users leave information signposts that help later users make sense of the wealth of alternatives available to them. Later users benefit from the signposts, because they are able to direct their attention to the parts of the site that are most valuable to them.

Three-dimensional on-line electronic environments allow people from globally distributed locations to “meet” (e.g. Activeworlds, Cyberworlds, see also Drozd et al., 2001). The character of such meetings can be anything from casual encounters between strangers in public spaces, through on-line games, to interactive TV or drama. The environments themselves differ with regard to the visual and physical qualities of the “material” structures they provide, the embodiments or avatars that users can employ, and the means of communication that are available: video, audio or textual chat.

Collaborative virtual environments, such as multi-user domains (MUDs), chatrooms, or three-dimensional graphical environments, provide a common space for people to interact in, independent of geographical location. In this chapter we examine how the different metaphors used to represent two- and three-dimensional environments might influence interpersonal behaviours. We focus on behaviours related to navigation and positioning: (1) proxemics — the maintenance of personal space, (2) the signaling of private space and (3) the effects of crowding. We discover that the design of the three-dimensional space offers sociopetal spaces that encourage interaction, make clusters of actors easily visible and provide cues so that people maintain a sense of personal space. In both environments, adverse reactions to crowding occur. We suggest that differences in interpersonal behaviours may be influenced by embodiment (avatar) design features of the space and the number of other actors present. In a three-dimensional environment, these factors appear to influence navigation and positioning in the environment.

This chapter outlines an approach to designing information spaces that we call experiential design, and illustrates the approach with examples of our recent work. The main virtue of this approach is that it claims to draw on universal primitives in the way people understand things, events, relationships — and information. And because of this virtue, it naturally supports social navigation of information spaces. The basic idea of experiential design is that, because we are embodied beings, meaning ultimately resides in bodily experiences. We have evolved to act in the physical world, and how we are able to understand abstract information is derived from that capacity. If we design for embodiment, understanding comes free; this is the first major benefit of the approach.

The basic idea behind location-based information systems is to connect information pieces to positions in outdoor or indoor space. Through position technologies such as Global Positioning System (GPS), GSM positioning, Wireless LAN positioning or Bluetooth positioning, the system keeps track of where a terminal (and its user) is located in space. Via his terminal, the user is allowed to enter/upload information, to which the system automatically allocates a latitude-longitude coordinate. Later, the same user, or some other user, can access that information (again via their wirelessly connected terminals) when they enter the place. Although the digital information is stored on a remote server away from the actual location, the position technology and the mobile terminals give users the impression that information is actually “attached” to the place where the user is. In this way, location-based information systems create user experiences similar to those of Post-its, graffiti and public signs and posters. In both cases, an information space is “superim posed” on indoor/outdoor space.

This chapter describes work in progress on a new way of approaching social navigation (Dourish and Chalmers, 1994; Benyon and Höök, 1997) through the use of populated, growing, knowledge gardens. These shared virtual landscapes provide an online space where people communicate and information can be “tended” through the affordances of an ecological metaphor. If we define social navigation as “finding things or going to places via, or with, other people”, and take that the whole process of categorising and finding information is a largely social process, then these Arcadian landscapes can provide a useful approach to social navigation in cooperative information applications.

The Footprints project was a multi-version, multi-year investigation into the application of social navigation principles to the problems of assisting web navigation. Inspired by Hill and Hollan’s original work on Readware (1993), we developed a theory of interaction history and built tools to apply this theory to navigation in a complex information space. The guiding principle was collecting, filtering, organising and redisplaying interaction history. We built a series of tools — map, paths, annotations and signposts — based on a physical- world navigation metaphor. This chapter1 presents results from the use of three versions of the system by a voluntary user group over two years and a controlled study with paid participants. Our user study showed that users who had the aid of Footprints were able to get the same amount of work done with significantly less effort in a given period of time.

The web is a social phenomenon, with businesses and individuals finding myriad reasons to both publish and access information. However, though millions of people are using it at any given moment, the web is a lonely place — much more like a catalogue than a bustling downtown. Users have individualised experiences that hide the presence of others. People think of the web as a kind of physical space in which they move to obtain information (Maglio and Barrett, 1997b; Maglio and Matlock, 1999), but they do not think of it as a kind of place that affords social interaction (Dieberger, 1999). Erickson (1993) defines “place” as “space plus meaning”, and Harrison and Dourish (1996) define it “as space which is invested with an understanding of behavioral appropriateness, [and] cultural expectations” (p. 69). Put simply, a sense of place derives from a shared understanding about a space, and interpersonal interactions are required to create shared understanding (see also Munro, Höök and Benyon, 1999). To transform the web from a simple space into a social place requires the structure of an interactive culture

In the early 1980s, researchers working in the area of interactive systems became increasingly interested in the topic of cooperative work. Human beings are, after all, social animals, and most activities in which we engage are conditioned by and conducted in coordination with other individuals. We work collectively. However, until this point, the focus of interactive systems research had largely been on a single user sitting at a desk in front of a computer screen. The field of computer-supported cooperative work (CSCW) emerged in response to this focus. It drew attention to the range of concerns that lay outside this computer/human dyad but which fundamentally affected the nature of work at the computer, such as the social setting in which the activity took place, and the role that that the activity played in an individual’s collaborative actions. In the years that have followed, CSCW has legitimised this concern and reoriented HCl to take into account the social context in which work is conducted. At the same time, it has also had an influence on the development of everyday technologies for computer-based work; as the Internet has become a more commonplace computational phenomenon, so technologies such as workflow and groupware have moved out of the research laboratory and into everyday computational practice.

Future information systems will be populated information spaces. Users of these systems will be aware of the activities of others, and what information they find useful or not. They will be able to point out and share information easily and even guide each other. These systems therefore will be social spaces. People associate social connotations with various types of spaces. These social connotations raise expectations about appropriate behaviour, privacy, trust, and private or public space in these virtual communities. This chapter discusses the relationship between social navigation and spatial metaphors and how social connotations of spatial metaphors can influence social activities in virtual space. We discuss a number of open issues in virtual communities in general and social navigation in particular, that are related to social connotations. We also present a pilot study that indicates that social connotations are perceived differently in real and virtual spaces.

Two complementary schools of thought exist with regard to the basic underlying assumptions and philosophies that guide our research in information navigation and access. As with all of HCl, and indeed most of informatics, we can place theories and design practices based in objectivity and mathematics at one end of a spectrum, and those emphasising subjectivity and language at the other. The first school of thought sees itself as part of traditional computer science, rooted in models that encompass the individual variations of users and that are often derived from experimentation and observation in controlled conditions. Mainstream information retrieval, cognitive psychology and task analysis exemplify such a philosophy. Complementary views are held by those who hold the sociological and the semiological as primary, and consider that objective categorical models are insufficient to model the complexity of human activity. Collaborative filtering, ecological psychology and ethnography are examples here. The techniques and systems presented in this book do not all lie towards one end of this spectrum, but instead show a variety of choices and emphases. This chapter, however, focuses on theory firmly towards the subjective and linguistic end of the spectrum: tools to let us place, compare and design techniques and systems. Such theory is noticeable by its near-absence in the literature of this burgeoning research area. Here we try to redress the balance, aiming to build a more abstract and general view of our work.

This chapter presents two empirical examples of information navigation work in organisational contexts. They show that information gains its relevance to any individual organisational actor in proportion to how actors in other organisations use that information. In this respect, it will argue that information is socially organised. The chapter will then discuss what implications this has for providing new sources of information and new information retrieval techniques with the World Wide Web.

Over the last few years we have been exploring an alternative conceptual isation of human-computer interaction (HCl) that sees HCl as the navigation of information spaces (Benyon and Höök, 1997). As a corollary cognitive engineers can be seen as the creators of information artefacts (Benyon, 1998b). The work is closely allied to some interesting developments in HCl, notably the concepts of “social navigation” presented in this volume and in (Munro, Höök and Benyon, 1999), but also to the ideas of distributed cognition (Hollan, Hutchins and Kirsh, 2002). Elsewhere in this volume (notably in the chapters by Spence and Chalmers) the concept of navigation itself is explored. In this chapter we report on our work concerned with exploring how well concepts that have been developed in disciplines such as architecture and urban planning transfer to information spaces.

In this chapter we examine how people think about the information space of the World Wide Web. We provide empirical evidence collected in interviews with beginning and experienced web users to show that much of people’s conceptual experience of the web is metaphorical and under-stood through the process of conceptual integration. We argue that designers of tools for navigation and collaboration in information space should consider how people experience web space, including the natural tendency to metaphorically construe information space in terms of physical space.

Navigation is a fundamental human activity which, in the physical world, has been carried out since time immemorial for purposes as wide ranging as adventure, conquest and foraging for food (Lewis, 1994). In electronic information spaces the aims are not wholly dissimilar. For example, in a menu system (Figure 17.1) a user is seeking a theatre to attend this evening: she may initially have no idea what’s available, and is exploring to form a mental model of possibilities before making a decision: a trail (Field and Apperley, 1990) allows her to selectively retreat to a previously visited discrete category label. On the other hand (Figure 17.2 — see also colour plate 14) the information space may be multidimensional and continuous, and the engineering designer may be examining a two-dimensional prosection (Tweedie et al., 1996) associated with two parameters under his control and colour-coded to indicate the success (green) or otherwise (grey-scale) of any design defined by values of X1 and X2.