The talking trees in Lord of the Rings or The Wizard of Oz may be coming soon to a forest near you. A group of computer science, electronic engineering, and physics professors is developing a system of tiny programmable semiconductor specks that can work together to sense, compute, and network wirelessly.
The 1 cubic millimeter specks would be scattered or sprayed on things like walls or clothes (or trees). Each little speck would have built-in computing ability and be able to communicate wirelessly over several centimeters to link with the others to form a powerful computational network called a “specknet.” So if, for instance, you walked through a room that had been “speckled,” the computer could not only detect exactly where you are, but actually communicate with you.
The research consortium, based at the University of Edinburgh in Scotland and three years into what is expected to be a twelve-to-fifteen-year project, believes that specknets will someday make computing truly ubiquitous. They envision applications such as spraying specks on the clothes of medical patients to monitor their health, sprinkling specks in the forest to provide early notice of forest fires, spraying them in disaster zones to provide communications infrastructure, and sprinkling them in military areas to detect intruders.
While they have not yet reached the goal of specks as tiny as a grain of rice, projects at the 5 millimeter size (about the size of a match-stick head) are well underway. Each speck is programmable and has about 2Kb of memory. That may not sound like a lot, but if you have thousands interacting with each other, the potential computational ability is significant. And, the researchers note, it takes less power to transmit information over short distances than to relay it all to a central gathering point.
One prototype, called Prospeckz, used specks on a combination of a coffee table, a throw blanket on a couch, and a “radio table.” The coffee table was designed so that if a user picked a book up from it, the lights would automatically turn on. The “smart throw” could turn on an electronic device such as a television when the users sits on it, and the radio table could automatically adjust its volume to make it louder as the user moves away from it.
Another prototype demonstrated a “mood cloud.” In that experiment, a computer keyboard with speck-type sensors in it was able to monitor the typist’s heart rate and change the lighting conditions above the user from bright when he or she was calm, to increasingly dark and ultimately to flashing as the user’s frustration level mounted.
[bluebox]
LETTER TO THE EDITOR
I have just read the article “Analyze This: A Task Analysis Primer for Web Design” in UX magazine [Vol. 5, Issue 1, 2006] about the value of task analysis (TA) in web design and found it very interesting. It is refreshing to read about practitioners encouraging the use of task analysis in a profession where it seems to have gone out of fashion.
I, too, am surprised at the lack of use of task analysis by usability specialists and the way many actually look down their noses at it. I agree with your reasons as to why task analysis is not done but I believe there are a few others that are important.
- The applications and uses of a task analysis are not effectively communicated: many people believe task analysis is the start and end of the analysis. In itself, the results of task analysis are effective in understanding the tasks that a user performs, but this only has value to the usability practitioners themselves. Managers, engineers, and customers don’t care that one task is performed after another or that it involves a set of subtasks. From my experience, the real value of task analysis, beyond understand-ing the task context, is the results from the secondary studies that can be performed using the task-analysis data such as human error analysis, workload assessment, task complexity assessments, heuristic interface evaluation, and training needs analysis, or for UI design guidance.
- There are people who develop overly complex and time-consuming methods for task analysis. This acts as a barrier to performing the analysis, firstly because it is difficult and time-consuming to do in relation to the benefits obtained from it; and, secondly because it is difficult to communicate the outcomes to engineers, managers, and customers.
- Task analysis is taught using examples such as “how to make acup of tea.” The reason for this is because it is a well-known task. However, it does not illustrate the benefits of doing the task analysis in the first place and does not illustrate the applications of the analysis once it has been done. It also belittles the value of task analysis when attempting to sell the approach to engineers.
I am sorry about the rant but it is a long time since an article has been written about the benefits of TA especially in the area of web design. Thank you for writing it!
[/bluebox]
