Metaphorical Design in Virtual Worlds
Introduction
Metaphors are key to creating affordances between a user and representational objects in computer systems. In this instance, metaphors are the conceptual tools which allow users to interact with a computer more efficiently–much has been made of the visual desktop metaphor, which enabled an untold number of novice computer users to make sense of how a computer can be used, organized, and eventually fit seamlessly within everyday life. However, metaphors can be used poorly, hinder user experience, and ultimately confuse the user, thus preventing affordances from forming. 3D virtual worlds rely heavily upon metaphorical design, but how effectively are the metaphors used? This paper will examine several examples of metaphorical design present within Second Life, the extent to which they rely upon and create affordances, and prospective directions for future designers of virtual worlds to pursue.
Literature Review
Existing research on metaphors, affordances, and frustration in HCI establishes the significance and value of evaluating each of these components in academic and practical design circles (though whether these concepts are put into practice is a different question). The links between each of these concepts are also well established, and this paper will attempt to put these theoretical principles within the context of virtual world computing, and how its unique challenges and opportunities impact and shape design mechanisms for future developers and researchers.
Metaphors, within the context of this report, are the essential building blocks of a 3D virtual environment. Metaphors have long been established as significant tools for user-centered design, dating as far back as the development of the desktop in the early 1980s. Indeed, a substantial amount of all visual operating systems depend upon metaphors to communicate their functions, and continue to evolve: in fact, as this researcher works on her paper, she used the “corkboard” function of her writing application to organize symbolic notecards with metadata for each segment of the paper. Many metaphors in computer systems have themselves become naturally intuitive, as long term users and younger generations feel increasingly at ease with menus, toolbars, buttons, stacks, docks, and more conceptualizations. Clearly, the development of the metaphor in computer systems is one of the most crucial aspects of human-computer interaction, and was a key enabler for new computer users.
However, metaphorical design is not in and of itself positive or helpful, and the history of HCI is littered with bad ideas and good intentions. One of the classic examples of inappropriate use of metaphors in design is Microsoft’s “Bob” application, designed to make users feel “more at home” with a personal computer. “Bob” was a “2 1/2”-D environment–a 2D environment designed to give users a sense of depth, and depicted a cozy living room with a fireplace, animated pets, and furniture. Unfortunately for Microsoft, “Bob” ultimately became a failed metaphor, since it created user frustration and dissatisfaction. This example illustrates briefly the power of a metaphor to either mitigate or create user frustration, which in turn has significant bearing upon the affordances users employ when negotiating computer systems.
Affordances can be generally defined as the properties of an object which make its function and meaning implicit to the user, and is a phenomenon which is born from interaction between the two. Essentially, it is how users instinctively know a button can be “pushed,” and how to manipulate objects in a virtual world. Consequently, if an object presents no affordances to the user, then an object’s implicit meaning must be learned.
Depending on a variety of factors–including learning aptitude, environmental factors, and system design–this learning process may be either productive or a source of frustration (or both). Within virtual world design, users must orient themselves within a novel environment; as new abilities and experiences are discovered and learned, users can take advantage of the new affordances to apply to unfamiliar activities. Appropriate metaphors employed by the designers make this process of learning, internalizing, and applying affordances less frustrating and more productive.
Log Overview
The logs included in this paper’s analysis focus on the following tasks:
Frustration present within the logs is predominantly caused by computer performance or lag. Learning tasks–shopping, seeking information, navigation or observation–don’t reflect any particular frustration with the task. User is moderately experienced with virtual worlds, and by comparison, Second Life has a long way to go. Controls are awkward, performance is frequently slow, and seeking objects or information in a crowded environment takes much longer than it would in other worlds. Some crashes are experienced by the user as well, which directly contribute to an increase in frustration levels. However, only two exercise logs indicate significant peaks in frustration and irritation levels, which are the Basic navigation log and the Sloog account log. Other exercise logs tend to have very low frustration levels.
The logs reflect some interesting metaphorical activities present within Second Life, such as “window shopping,” “opening objects,” “touch,” and “teleporting.” As a companion to this, many objects observed within Second Life are grounded in metaphor, such as notecards, landmarks, and shopping bags. However, many objects and actions in Second Life break away from established metaphors, or behave unpredictably; these can cause initial frustration on the part of the user. Phrases such as “equip” or “attach” give no indication of how to accomplish a certain task, and rely upon user experimentation or prior experience to determine a working solution.
Also present within the logs are some of the user’s expectations of how the system will work, how efficiently a task will be completed, and how relevant and enjoyable a task will be. Consistently across all logs, the user has high expectations of success in a task, low expectations of how relevant a task is to career objectives, and low expectations of how Second Life will facilitate completing a task. “Enjoyability” of a task is variable within the logs. The measurable confirmation or violation of these expectations are evidenced within the affective responses the user reports; these reactions range from expressions of relief and surprising enjoyment, to boredom and disengagement with content.
Evaluation Criteria
Criteria for evaluation of metaphors is based upon the following:
Levels and causes of frustration, measured within the logs.
Frustration levels are measured on a scale of 1 to 10, with 10 being incredibly frustrated. Throughout the exercises, frustration levels remained consistently low (a score of 3 or below), with the exception of the initial exercise at Virtual Ability Island. In this instance, the logs reflect frustration levels increasing from 3 to a peak of 7, with a sudden decrease to 2, then 1. The reason for this frustration is related to computer performance–once the user switched from a relatively low-powered laptop to a desktop computer with a dedicated graphics card, frustration levels decreased substantially. This correlation provides significant clues to design flaws and metaphorical challenges within Second Life.
Relative time to complete a given task.
For the most part, task completion time is not specifically recorded in the logs, though some language present in the logs indicate whether the task is proceeding slowly or quickly. Often–though not always–this corresponds to frustration levels on behalf of the user.
User expectations–confirmed, unmet, or violated
User expectations are key to understanding the effectiveness of metaphorical design. Expectations can fall into three categories: an expectation is confirmed when an object or interaction behaves in a predictable manner; an expectation is unmet if the object is non-responsive; and an expectation is violated if the object behaves in an unpredictable manner. Throughout the logs, expectations are consistently confirmed, unmet, or violated by Second Life objects.
Expectation adjustments
Over time, the logs reflect decreasing levels of frustration without significant changes to either the hardware used or Second Life. This signifies an adjustment of the user’s expectations of efficiency, speed, or difficulty of a given task. With this information, we can posit that the user is making new connections to the interface, which can be simultaneously beneficial and problematic for design metaphor study.
Virtual Experience Metaphors
Second Life as an interface relies heavily upon metaphorical representation of objects, actions, and concepts present in a user’s physical world. This section will examine metaphors present in three areas: menus, environmental objects, and action representation.
The very concept of a menu is itself a long-established metaphor in HCI, representing a selection of choices for user action. Menus in Second Life are both traditional drop down style, and a contextual circular style. The former of these menus offers several metaphorical options–changes in lighting, setting a virtual location as “home,” adding to an avatar’s bank account, and taking a snapshot. Some of these metaphors are successfully employed–”home” implies a safe place which a user can return to instantly, which is exactly how the function works, and which was immediately understood by the user in the “Getting started” exercise. Controlling visibility via ambient lighting is also reasonably successful, since the light levels at certain times of day are instantly recognizable and utilize affordances already familiar to a user. On the other hand, many of these metaphors are poorly executed, and ultimately can cause frustration to a user. Notorious among these is the “Inventory,” where all an avatar’s virtual objects reside; determining how to interact with ambiguously named objects is a consistent source of frustration to users. Additionally, while the Inventory has some categorization built in via its default folders, this quickly falls by the wayside, and objects become increasingly difficult to identify and locate. Management of inventory items quickly becomes a time-consuming chore, and user logs reflect growing irritation with seeking ambiguously named and labeled items.
In addition to menus serving as a core metaphor for using and navigating Second Life, much of the environment is made up of symbolic or literal metaphors. The most significant of these is the avatar itself, which is a metaphor for the user’s actions within an environment. Indeed, with the addition of gestures and facial expressions, an avatar can communicate a wide range of emotional responses, and engage in a “local” conversation with other avatars. Other core metaphors to the Second Life experience include notecards, which are primary communication tools intended for wide distribution, and landmarks, which are a set of virtual coordinates for a sim. These metaphors are put in place as systemic conventions by the Linden Labs designers, but the unique and powerful aspect of Second Life is the ability for any participant to become a developer in the environment. Many user-created metaphors encountered in the exercises include newsstands, which distribute notices and magazines to users; ATM machines for avatars who have Linden dollars associated with a virtual bank or store; and objects shaped to look like gift boxes or shopping bags, indicating their need to be opened. While completing the formal dress task, the user reported easily understanding the significance of such objects, and low levels of frustration were reported–this is a fine example of affordances facilitating action. While some of these metaphors are successful, others are not–many objects have ambiguous meaning, do not respond to touch, or create anxiety in the user. One object the user encountered while exploring was called “fireflies,” which the user tested at her home location. This object simulates the movement and glow of lightning bugs, and the user assumed this would be generated simply by dragging it onto the ground. To her dismay, the object needed to be attached to an avatar or another object, otherwise it would be extremely difficult to stop or select–this increased the amount of time it took to complete a task, and added to her anxiety. This is one example of an insufficiently explained metaphorical object causing frustration to a user, but many others can be found.
Finally, action representation within Second Life is typically entirely metaphorical in nature. Avatars can “touch” or “buy” objects, create new objects, and assign individual objects to perform specific actions with scripting. These examples have fairly obvious expectations attached to them; when an object is touched, users expect the object to respond in some way, either by triggering a script, or presenting the user with a menu. Buy is also relatively unambiguous, as are sit and stand. More problematic metaphors are the “attach,” “detach,” and “drop” functions–in order for an avatar to drink from a glass, carry a purse, or even wear more realistic hairstyles, these objects must be attached to the user. Typically, human beings do not “attach” drinks to themselves, so this metaphor is an unusual choice–”hold” or “put on” might be more appropriate. Similarly, “detach” and “drop” are two virtually synonymous terms which have two very different effects–detach removes a worn object and returns it to inventory, while drop literally drops the object on the ground. These terms cause unnecessary confusion, which is reflected in early user logs.
Conclusions
Areas for Further Research
In order to form more concrete links between metaphors and affordances in system design of virtual worlds, other researchers should study user expectations of behavior, as well as the affective responses users have to certain objects in virtual worlds. Expectations are crucial to understanding the effect of a metaphor, as well as the presence of affordances–particularly useful would be studying the violations of certain expectations, their positive or negative impact, and how users accommodate those violations.
A topic of study outside the scope of this paper would be the convivial aspects of metaphorical design within Second Life. To what extent can metaphorical objects–such as a forum or voting machine–spark convivial participation within a community? Are the objects already employed in certain communities, and what effect do they have upon community discourse and interpersonal connections? To what extent does conviviality translate to activism? This avenue of study has significant importance to the study of digital government within virtual worlds, but applies to any study of social networks and the artifacts which enhance their formation.
Design Concepts
Overall, virtual worlds present unique and unprecedented opportunities for both system designers and virtual world inhabitants to translate metaphorical concepts into concrete actions. While successful metaphors are often difficult to explain, some basic criteria exist to evaluate the success or failure of certain metaphors within virtual worlds. Designers would benefit from user testing of such metaphors with a diverse group of users to determine which metaphors most successfully allow a user to create or employ affordances.
Ultimately, though, affordances are about intuitiveness–our collective human experience tells us what to do with a pen, a light switch, or a book. However, as a new generation of users becomes so accustomed to the presence of technology, intuitiveness will take on new dimensions in computer systems, and that which is difficult for users today to grasp will be natural to them. Designers must be sensitive to the transient nature of affordances when creating metaphors, and ultimately work to reconcile the complexity of virtual worlds with the simplicity of affordances.
References
Bærentsen, K., & Trettvik, J. (2002). An activity theory approach to affordance. Proceedings of the Second Nordic Conference on Human-Computer interaction, 31, 51-60. Retrieved April 1, 2009, from http://doi.acm.org/10.1145/572020.572028
Blackwell, A. (2006). The reification of metaphor as a design tool. ACM Trans. Comput.-Hum. Interact., 13(4), 490-530. Retrieved April 1, 2009, from http://doi.acm.org/10.1145/1188816.1188820
Preece, J., Rogers, Y., & Sharp, H. (2007). “Affective Aspects.” Interaction Design: Beyond Human-Computer Interaction. New York, NY: Wiley. Retrieved January 29, 2009.
About this entry
You’re currently reading “Metaphorical Design in Virtual Worlds,” an entry on Kate Dohe
- Published:
- April 25, 2009 / 10:56 PM
- Category:
- Library
- Tags:
