Sunday, October 17, 2010

Ubiquitous computing that has an impact on people

(This post is in response to Rogers Y. 's Moving on from Weiser's Vision of Calm Computing: Engaging UbiComp Experiences.)

Last year I met Prof. James Landay at MSRA and I was quite impressed by two of his projects - UbiGreen and UbiFit. Essentially, what they did was using sensors to collect data from users' daily activity such as walking, biking, taking a bus, etc. and then showing how this data relates to one's healthy habits or environmental friendliness.

I truly love research that has an impact on people. Another such example is the mobile learning project by Prof. Matthew Kam in CMU (technically this does not fall into the category of UbiComp).

Having read Yvonne Rogers' paper, I believe what she is concerned about is which direction of UbiComp is more promising. By saying 'promising', it is implied that there are some certain criteria judging, for example, Weiser' view.

1. Review: what is calm computing?
In my opinion, the spirit of calm computing is about the way we arrange information complexity, making it helping and serving instead of distracting or even jeopardizing our activity. Yvonne Rogers might not actually mean this; she points to a wider concept of Weiser's view of UbiComp.

2. The taxonomy of UbiComp
As Yvonne Rogers points out, and I paraphrase, basically there are two approaches toward ubiquitous computing. One is trying to make our surrounding smart and reactive, relieving people's mental and physical labor. The other is making the system portable and using it to get people engaged in certain kinds of activities.

3. The critical constructiveness of the paper
The author of this paper strongly supports the later approach, citing three areas of study that already made a difference in UbiComp research. However, she has not explained clearly why she supports that approach. Even though she mentioned she thought state-of-the-art results were limited and not successful enough, this might just be a motivation of turning the wheel not the reason. For me, I favor more the latter approach and the reason is given at the beginning of this post.