Panel 3: Data Visualization and Device Prototyping

Jeff

Jeff mostly talked about his findings while trying to design a thesis searcher/browser for Stanford. First of all, I noticed that the data visualization he tried to do went through several drafts/stages. He started out with a gray-scale matrix, admitted that it is definitely not very human friendly to look at, and moved on to demonstrate how they improved on the visualization. Second of all, he showed how data visualization can tell the wrong story and make people draw the wrong conclusions. In particular, I found it interesting how he used a form of data visualization (with the discipline in the center, and related disciplines surrounding it like a circle) to show another form of data visualization was flawed. I also like how he used his colleague as a source of judgement on the quality/effectiveness of his data visualization, and was able to find out about certain assumptions he made that was false (namely, when questions aren’t symmetrical, and “how similar is biology to chemistry” is different than “how similar is chemistry to biology”).

Kristine

Kristine talked about her experience prototyping EMAR, a robot that’s meant to access mental health of students in real time. I especially liked how she put together the first design, and how she had to throw in some scrap cardboard from her house at the last minute. She talked about how scrapping is surprisingly quick and dirty, and how it still gets the job done. Personally I would be a little uneasy about having to prototype an entire robot, but Kristine showed how she was able to do it. She sketched designs of the robot, including size and looks. She came up with a spec, detailing what she wants the robot to be able to do (such as say hello, ask questions, etc). One thing that stood out to me was that she talked about the step after device prototyping. Kristine talked about showing the prototype to users (high school students, in this case) and asking for feedback. Using that feedback, she was able to update some of the robot’s functions and looks (such as implementing blinking eye to make it look more human and appealing to students).

What now?

From what I learned, I can use it to improve my project in several ways. For my data visualization project, I should sit down and decide if the data visualization actually answers the question correctly. I wasn’t really considering the idea that data visualization can be flawed, so I need to look out for that aspect. For my device prototyping project, I should show it to potential users and have them actually test it out. This way, I can get feedback on how easy it was to set up, how effective it was, and what areas I would need improvements on. I should also be more organized with device prototyping, and come up with a spec so the function of the device would be clear.