Shallow & Wide vs. Deep & Narrow

There was an interesting work at HCI Korea 2010. When a family or friends play a video game together, they enjoy not only the game itself but also each other’s reactions and facial expressions. So, this work videotapes the people’s faces and behaviors at important moments while they playing a game. There could be many ways to decide “important moments”, and this work uses sound signals from the game. When I talked to one of this work’s designers, he was quite humble to say, “It is not a fascinating technology.” I think the main idea is great, and “simply” using audio signals from the game is a wise choice. But he continued to say, “We do shallow and wide research.”

It is difficult to choose between shallow & wide research and deep & narrow research. Of course, deep & wide is the best, if possible. The importance of interdisciplinary and hybrid research has been emphasized. I agree on its usefulness and values, but I’m not sure about the effectiveness of starting one’s research from those hybrid fields. To be honest, I think a grad student should study deeply and narrowly first, and start working on applications and collaborate with people from other fields once his foundation has become pretty much hard. Looking at the MIT Media Lab case, I feel like they bring up brilliant ideas and combine various technologies into real world applications, but each individual is an expert in his/her own field. For Dr. Cha also, although she’s now doing high-level research with Twitter, very closely related to sociology, her background is about low-level networking systems.

I don’t know the answer.

Shallow & Wide vs. Deep & Narrow