In the article “The Visual Rhetoric of Data Displays: The Clarity Conundrum”, Kostelnick dissects graphic design’s application in to data displays. For this, I will be mainly focusing on his first explanation, which is the application within science. He primarily used the word “clarity” as his focal point of this section. Clarity is the extent in which the data itself can reach the audience and how well and clearly this information is interpreted. To further describe this term, he explains that clarity is more significant for statisticians and researchers, who are focusing only on the data itself, as opposed to the graphic designers who are producing the graph and displays. This is because graphic designers are primarily concerned with the audience’s attention and perception. He then identifies Tufte specifically about their stance on data display. He attacks their methods of cropping down the data in order to make the display more simplistic and appealing to the audience and only leaving what is absolutely necessary to interpret the data. Kostelnick claims that this crops out important sections of the data that help the audience accurately submerge themselves in the data. One thing he agrees with them on is their idea of “chartjunk”. He seems to be equally as critical, but he also qualifies it as being sometimes appropriate, that it depended on the kairos.
On Kostelnick’s stance on scientific data visualization, I feel that he may be a bit too harsh on the negation of information. He thinks everything holds at least minor significance in reception of the data, which may be a stretch. I think that it may play its part and have a purpose, but I don’t think it necessarily takes from the audience’s ability to clearly understand the information. Although, I definitely appreciate his explanation of chartjunk and its connection to kairos, because sometimes the data truly speaks to itself and chartjunk is necessary in the context it’s displayed.
My question is, what are some scenarios in which chartjunk may be useful and when does it take away from the data? Why? Kostelnick gives a few examples, but can this be elaborated on?