Monday, November 3, 2008

Ubiquitous Computing Still Hasn't Disappeared

Abowd and Mynatt's (2000) fascinating and prescient article, "Charting Past, Present, and Future Research in Ubiquitous Computing," offers insight not just for ubiquitous computing (ubicomp) researchers but anyone with an interest in technology design for "everyday living," as they put it.  Their essay examines ubicomp work during the 90s and offers a series of useful guidelines for thinking about ubicomp in context as well as recommendations for future research. 

The remarkable diffusion of computing into our physical world represents more than just easily available technology, rather, "it suggests new paradigms of interaction inspired by constant access to information and computational capabilities" (28).  These new interactions are easy to spot, especially in the last few years as mobile platforms like the iPhone have changed the way many people relate to their data as well as how they act in social situations (i.e., taking calls when with a friend, texting while someone is trying to talk to you or my favorite: The play-with-my-phone-to-avoid-the-pain-of-this-awkward-silence game). 

So, how far have we come since 2000?  Abowd and Mynatt argue that, "current systems focus their interaction on the identity of one particular user, rarely incorporating identity information about other people in the environment.  As human beings, we tailor our activities and recall events from the past based on the presence of other people" (37).  I agree with the statement, but times have changed.  We are now deeply involved in the context of others
, though often it can be construed as superficial.  Social networks like Twitter and Facebook allow us to broadcast our feelings and daily adventures.  However, is this what the authors had in mind?  Though we have made much "progress" in blending others into our digital lives, most of this information is focused on the present, and in the case of Twitter, it is the micro-present.  Our dominant, seemingly ubiquitous social networks are designed for the present.  Twitter, in particular, is designed to encourage micro-updates of 140 characters maximum. 

Much of our technology is designed for the now
, but where is our past represented in the digital ecology?  I argue that we do have an astonishing digital past in the form of email and instant messaging archives.  (text messages are not saved to a central server by default, so they seem, unfortunately, to live in a state of constant disappearance.) Arguably, there has never been a time in history when the historical record has been more complete or rich.  People who would never write thousands of letters do write as many emails, and they are (hopefully) preserved in archives.  The question I have for ubicomp is this: How can we design ubiquitous devices and software to somehow harness the power of our rapidly growing personal archives?  In other words, how (or even should we) incorporate the past into our ubiquitous digital lives?  What could we learn about ourselves if we had a way (if we choose) to harness and easily visualize our past using ubicomp techniques?  The mobile phone seems to be the key.  What if you bought a pack of cigarettes and instead of your phone showing a waring that smoking kills, it popped up with a quote from an email you received three months earlier from your wife that said "Please don't smoke.  I don't want you to get sick."  This would incorporate our present environment, our past and our future health all using ubicomp.   

This particular scene can be usefully broken down using Abowd and Mynatt's five-point framework for thinking about context in ubicomp: 

  1. Who: “Current systems focus their interaction on the identity of one particular user, rarely incorporating identity information about other people in the environment" (37). My scenario brings the user and other people in the user's life into context.  Although it's not a "real-time" interaction with your wife, why does it have to be?
  2. What: “The interaction in current systems either assumes what the user is doing or leaves the question open” (37). With GPS enabled phones and the emerging use of mobile phones as credit card devices, the device will not have to assume what you are doing.  It will know where you are and what you bought (of course, you could just pay cash).
  3. Where: “In many ways, the 'where' component of context has been explored more than the others" (37). Obviously, GPS finally solves the problem of “where.”  The key to understanding the importance of “where” depends upon how well our ubiquitous technology appears to us at critical moments and steps in to help.
  4. When: “...most context-driven applications are unaware of the passage of time” (37).  Linking context to time is crucial for developing truly aware applications.  In my scenario, the mobile device could use the time of day and GPS to send a warning before the user buys cigarettes.  Say, for example, if it was two o'clock in the morning on a Friday night, and the user enters a convenience store.  The device, based upon a baseline of past activity, might try to warn the user not to buy cigarettes.
  5. Why: “Even more challenging than perceiving 'what' a person is doing is understanding 'why' that person is doing it" (37).  Trying to ask “why” a person is doing something does not, to me, seem like a fruitful question for computers to ponder.  Instead, humans should ask these questions about themselves.  However, in my scenario, the computer simply prompts the user using the emotionally charged form of personal email to facilitate reflection about why they are doing what they are doing--right now.  This seems to me the best use of ubicomp and computing in general:  Rather than giving answers, computers should ask better questions and let humans do their own answering. 
    As the authors note, HCI tends to design for closure, but everyday computing believes that daily activities "rarely have a clear beginning or end" (43).  This is a critical observation.  Life ebbs and flows, and our technology ought to accommodate our human reality, not constrain us inside of a designer’s assumption box.  

    Mark Weiser (1991) wrote that, “The most profound technologies are those that disappear. They weave themselves into the fabric of everyday life until they are indistinguishable from it” (1).  This is profound guidance for all designers, not just ubicomp.  My sense is that today, too much of my technology is in my face, so to speak.  I want my technology to quietly fade in when called upon, and I want it to leave me alone unless I need it. 
    The ironic challenge for ubicomp is not to make more stuff but to make more stuff disappear. 


    Abowd, G. & Mynatt, E. (2000). Charting past, present, and future research in ubiquitous computing. ACM Transactions on Human-Computer Interaction, 7(1), 29-58.

    Weiser, M. (1991). The computer for the 21st century. Scientific American, 265(3), 94-104.

    1 comment:

    Anonymous said...

    I like the challenge to have our technology be faded in the background, but feel that it is destined for doom. Could it ever be that clear, even if we aren't techy, the world that surrounds us is.