Reading 2 - Subtleties of Color

The blog explores the problem of color in data visualization, when most often a color palette is picked out with free whims and contaminates the data set.

One interesting point in the beginning of the blog is the inaccuracy of human's perception of color with the computer generated gradient. It points out that among the 3 main variables, lightness is the most common and easiest one to spot. There're 3 main color palette put in use, sequential data palette, divergent color palette, and categorical color palette. The one thing I haven't noticed before is the accessibility of the colors. When color blindness is taken into consideration, the choice of accessible color palette can be come very challenging.

At the end of the blog, we've also get a sense of color's meaning is more or less cultural and relies heavily on context (the data set). When the color can intuitively communicate the data's information (eg. green as plants, grey as deserted lands, blue as water), it's highly recommended to follow such rules to help the general readers understand the data more easily.