Monday, September 5, 2011

Reaction: Graphical Perception: Theory, Experimentation and the Application to the Development of Graphical Models

Overall this paper was an enjoyable read. The paper discussed a couple of types of charts I am unfamiliar with and I enjoyed reading more about them (these were the curve-difference chart and the triple scatter plot). The author discusses towards the end the difficulty in getting test subjects that will read a chart during the test in the same way that they would normally. For their tests they 'dumbed' down the charts, removing some elements to make it harder for the subject to perform mental division. While I understand why this is done, it would be interesting to try the same study without those subtractions. Also I wonder if there are ways to get the test subjects to behave more normally, and how other studys have attempted to compensate for the discrepancy.

My biggest complaint about this paper is in its physical design and layout. The decision to not put charts and graphs on the same page, or in some cases even near, the descriptions and critiques of them forces the reader to constantly go back and forth between pages. In some instances a reference to a particular chart came 4 pages before that actual chart. This detracts from the reader, as I found myself having to constantly reread information due to being forced to flip back and forth so much. While this is most likely made worse by being a digital document, it would be similar even in paper form. I did not previously realize how important the distance between a reference to a graph and the actual graph itself can be.