Reactive Environments
I think that the light switch described in this paper is only smart in certain instances. There are definitely times where it’d be problematic having a manual on-off that reverts to motion detection when there is no longer motion. The most obvious one is watching a movie with a group. If you turn the switch off to watch the film, and then one person gets up for some reason, the lights would turn back on, disrupting the film. Another example is sleeping. If you are tossing and turning in bed, the light switch might turn on and wake you up. Personally, I’d prefer a light switch with 3 settings, motion detect/on/off. While problems may be encountered when the switch is turned manual off, and less power would be saved with the switch in manual on, those problems seem to be less than the numerous inconveniences with a light switch that cannot be forced off.
Interface note: the format of this HCI paper does not seem to notice the HCI that take place when reading it. If you are reading a paper and on the next page it uses a photo that doesn’t take up the width of the paper, intuitively the photo should be on the right so that your eyes look to the left and see the text. Instead, many of the photos are on the left at the beginning of the page, causing jarring confusion each time one is reached. Furthermore, the transfer from page 5 to 6 on this paper combines a jarring placed image, a word broken over two pages and a second image 4 lines below the start of the text on the left side, making the paper hard to read.
How does one implement these daemons that monitor actions across the technology in a world that is constantly updating its technology. If we were still living in 1997 and using the same VCR, I can see how we could refine the interactions between technologies in the Reactive Room so that they were invisible. But instead we live in a world that is constantly updating technology, replacing hardware, replacing companies! Not to mention the standards between those technologies, not built with one another in mind. It seems, to practically have an environment like the one in the Reactive Room, what first would need to be established is an interfacing standard across these technologies so that different pieces of hardware can be connected together without having to rewrite the rules of interaction every time a piece of hardware changes. I might be dreaming though.
Proxemic Interactions
Mobile tokens are an ingenious way of handling interactions with a system. They deal with issues faced with hands being used to interact with others and interacting with the system. While this works very well for me conceptually, I’m still a bit torn on some of the implicit interactions discussed in the media player. I’m not sure I’d want my movie to pause if I was looking away or for the volume to go down if conversation arises. For some of these implicit interactions to work well, I think there will need to be an initial explicit interaction to configure them in the best possible manner.
