Paul-A-R02

Envisioning Information

Layering and Separation

This section dealt with reducing noise in a data graphic through the methods of layering and separating visual elements so that complex and detailed data could be clearly represented. Many of the suggestions for good data representation correspond to techniques employed in all visual art. The theories of Joseph Albers in regard to the effects of juxtaposed colors, lines and shapes were quoted several times, usually in regard to the negative effects they introduce into a data graphic in the form of noise. Bright colors that visually interact with each other are frowned upon because they create visual artifacts that distract from the data. However, the use of a small amount of highly saturated color in an otherwise “quiet” color scheme is actually recommended and considered one of the design secrets of classical cartographers in highlighting data detail. The use of subdued colors is promoted as a good way to define objects in a data map without distracting from the data content.

Another method for layering data is the value or lightness/darkness of a line or object. Gray staff lines work well for musical composition for example since they place the emphasis on the musical notation. The rule of thumb seems to be that the more contrast there is between the data and the ground that it is placed on, the less noise there will be. Conversely dark lines and fields on a white background are to be avoided. I especially liked the statement that “Dark gridlines are chartjunk” not only because it was informative about the effect of distracting lines in the presentation of data, but that the author seems to feel so strongly about it that he created his own derogatory term to describe it. His suggestion to eliminate grid lines in charts and tables wherever possible would best be taken to heart.

The discussion about negative space in a graphic design was very informative. Empty areas can draw our attention because they provide a contrast to the other areas of the graphic image. They can also provide an area where other objects will visually interact. The example was given of a grid of heavily framed boxes with spaces between them. At the corners of the boxes flashing circles appear as a visual anomaly that will create a strong distraction to the data presented. Similarly, putting boxes around data creates a negative space in the area between the box and the data leading to noise. The use of negative space can be a powerful tool in the physical presentation of data. The graphic examples of objects that are implied by the objects that surround them is very instructive. I wonder if the same could be done with the data that is presented. If a graphic were made presenting data on the American attitudes toward the world’s religions and some of the major religions were not included, would that make a strong statement? I think it could if there were some way to show the gap.

Small Multiples

This section was not clear to me at first. I could understand the train light example with the multiple images of the train and different light configurations for comparison. The Kanji writing example seemed more like a tutorial than a presentation of data. Perhaps I am placing limitations on my way of thinking due to the examples that have been used so far. It is a presentation of information after all. The same could be said for the Dighton Writing Rock example. I am not sure what the point of the comparisons are except that different persons represented the images in different ways. The time scale does not seem to be an important factor relating to the data. I would agree with the author that the comparison of the mountains of the world was not as well done as the rivers. There is no common baseline for the mountain ranges with which to compare their heights, something which would be of greater interest than their shape.