Photo courtesy of andyi

As part of the graduate course on Ubiquitous Computing the students were asked to comment on the idea of ubiquitous computing as "invisible" computing. Below is the response from Joel Ross, posted with his permission. Here is why Joel thinks invisible computing resonates with users, but is problematic as a concept. "In his paper “The Computer for the 21 Century", Mark Weiser proposed a vision of ubiquitous computing—a type of computing that was embedded in the world around us and that we would unconsciously use. Although the idea of “invisible computing” is very popular, Weiser’s vision has some fundamental problems that have keep his style of pervasive computing from becoming truly ubiquitous. Weiser’s idea of “invisible” computing resonates with users because people don’t understand (and so don’t really like) using computers, but is problematic because no tool can become and remain fully invisible.

Many people don’t understand computers. This was true when Weiser wrote his paper and remains true today. Computers are complicated machines capable of a vast variety of general and specific tasks and require a certain level of expertise to be able to operate. The “arcane aura” surrounding computers (and computing in general) discourages many potential users, and makes it so they don’t enjoy using a computer. Computing becomes an unpleasant chore for them, as they focus on the problem of using the computer rather than on the task at hand. Using the computer seems to take more effort than performing the job without it--the tool only gets in the way.

Thus Weiser’s vision of the “invisible” computer is very compelling. Many believe that a computer should be like a hammer—you pick it up and can use it without thinking about how it works. Weiser’s example of a car as a complex machine that can be used without thinking resonates with users—people learned how to drive a car, why is it so hard to learn to use a computer? Finally, the language of Weiser’s vision implies a very desirable world, in which using a computer (and thus doing everything that a computers can do, which most users will agree is “a lot”) is as easy as reading a billboard or a street sign. The implied power and capabilities are staggering: computers are capable of amazing things, so being able to do those things with a computer with no more effort than reading a few sentences is certainly desirable). Thus it’s easy to see why the idea of “invisible” computing resonates so strongly with users.

But achieving this vision is problematic, for no tool can become fully invisible. Just consider for a moment the idea of a literally invisible hammer—how could you ever use it? Dourish states that proponents of invisible computing need to find a better term, because the word “invisible” is misleading. When Weiser writes of “invisible computing,” I think he really intended to say “unconscious computing” (Dourish states that proponents of invisible computing need to find a better term). The goal is to reach a state that Heidegger calls “ready-at-hand.” When an object is ready-at-hand, it is being used as part of an action to complete a task, thus “disappearing” into the actions that the user is taking. This state is contrasted with being “present-at-hand,” when the user is aware of the tool as a tool independent from the user, the action, and the task. People can quickly move between these two states, in what Dourish calls “coupling.” Thus a person can be aware of a hammer as an object, yet quickly pick it up and treat it as an extension of his or her arm. Weiser’s vision is for computers to easily become ready-at-hand, but the idea he proposed (and the one that many people latch on to) is for computers to constantly become ready-at-hand: in other words, invisible.

This is a fundamental flaw in people’s interpretation of Weiser’s vision, for an object cannot be constantly ready-at-hand. All tools can draw attention to themselves, especially at the “seams”— when the use of a technology breaks down. While it may be easy to couple a hammer, that coupling will frequently be broken if the head of the hammer is loose and in danger of flying off. Similarly, even Sal’s world described in Weiser’s article would break down when the power goes out, drawing attention to itself and becoming visible and present-at-hand. This points to some of the ideas suggested by Mainwaring et al.’s paper titled “Infrastructures and Their Discontents: Implications for Ubicomp”—an infrastructure may seem invisible until it breaks down, or when it is the becomes an active focus (such as by attempting to avoid it). An infrastructure of embedded computers might easily become invisible (be easily backgrounded), but it would still never be fully invisible as some visions of ubiquitous computing suggest. The ideal of a computer just “working” all the time, to take as much effort to use as to read a sign (that isn’t smeared or far away), is impossible to achieve. No matter how easy a computer is to use, it will always be able to become visible.

However, it is possible for a computer to become easier to background. Tolmie et al. discuss the idea of an event or object being “unremarkable”—that is, something that is not remarked upon as being special. They point out that artifacts involved in a routine often become unremarkable, such as a knock on a door or the buzzing of an alarm. This implies that making computing routine can make it unremarkable. As Tolmie says: “a fundamental issue for us in things that are ‘invisible in use’ is not the physical nature or particular perceptual qualities of these things but rather the significance which accrues to them within a particular course of action.” Once using a computer becomes part of the action, it can be backgrounded (coupled, made ready-at-hand), and so in effect becomes invisible.

In this way, Weiser’s idea of invisible ubiquitous computing is not necessary distinct from normal “desktop computing”: using a desktop computer can also become unremarkable. For example, it doesn’t take me much effort to couple the movement of the keys with the text I am typing; the mechanics of browsing the Internet and checking my email occurs unremarkably in the background of my consciousness. I have learned how to use a computer the same way I learned how to drive a car, and just like when driving a car, it is possible to easily couple the use of a desktop computer with the task being performed. My usage of the computer in many ways has become invisible, even without a drastic shift to a different interaction mode. Thus achieving Weiser’s vision of invisibility computing may not require a major shift in the form of computers, just the ways in which people use them.

As people become more used to using computers (for example, by being part of a generation that grew up using them), I believe that computing will and has in many ways become more invisible to use (if not invisible to see) of its own accord. Much computing is becoming invisible, such as the use of cell phones and mobile devices. So just as Bell and Dourish suggest in “Yesterday’s Tomorrows,” ubiquitous computing is here already, if we only acknowledge its presence. "