Petit Response 1:
Before reading Look at data by Kieran Healy, I was getting curious about how ”colors” in data visualization were to set to function where the change of which is often executed in codes or provided as a pre-defined color palette these days. Personally, 1.3.1 Edges, contrasts and colors among this chapter shed some light on me regarding this subject.
It is rather easy to use a line of code to change a color’s attributes, mapping it to the change of a variable. However, this type of “numerical” mapping does not guarantee the same level of change in our visual perception. While working on the “clock” assignment, I realized not each and every alpha change yielded a same degree of difference. I also noticed that some colors showed more variance than others. I was glad to learn how this phenomenon was explained in a clear language of science and how we could retain perceptual uniformity within a certain colorspace, in this chapter.
However, while describing sciences and observations on visual perception, Healy also says the following, which I'd like to question.
“Different color spaces have been defined and standardized in ways….. Our decisions about color will focus more on when and how it should be used.”
Can we really say that color spaces have been defined and that our job does not expand there? I reckon that the perspectives in social science and data visualization must differ. Healy clearly leaves it up to somebody else. Then, this can be a task for me and my peers. I’d like to understand different color spaces’ psychological and emotional effect more clearly, before taking the science as proven and using colorspaces without giving enough suspicions.
Petit Response 2:
Despite its confusing graphics (bad taste!), Nigel Holmes’s “Monstrous Costs” somehow became a more memorable representation of data, compared to other charts following conventional routes. It was successful, perhaps because human beings are emotional, susceptible to creativity and novelty. A proven theory cannot always explain a new phenomenon.
Have something to say. Be honest with data. Think clearly about graphs. Use elements that work for people and data. After that, we’ll have to find a way to make the delivery of information creative and memorable enough. In this information overloaded era where a lot of things seemed to be already given as a science, it is ever more important to set our goals (in the data visualization) to understand human psychology and their perceptual limits and find a way to reach them in each specific context. I want that to be one of the messages I took from this reading.