All of this we have learned as a freshmen in Pratt in LCD class. However, It was interesting to read about color theory through the scope of data visualization. Talking about in terms of cartography and how color palette has to fallow basic and also has to fallow cultural associations and it has to work intuitively.
sequential palette is to show step by step gradient change. This palette is to picture gradual info changes. the reading mentions upsides and downsides of grey scale usage. They also say when you use color it is good to have two different hues to change gradually. a lot of color platforms can have short color range for certain colors.
Divergent palettes are useful when there are two different end of values that the user wants to compare to. This is why it is important to make sure the middle value color do not have any association with two ends of the palettes. Also, palette makers has to be careful about choosing color for two different ends. this is because color blind person cannot differentiate some colors. As a person who lays out info, people has to make sure all people can read your informations.
Qualitative palette is to convey difference between random values. tip to making this pallet is to avoid similar colors and usually colors under seven.(because when colors go above seven, it is easy to end up in colors that overlap each other). In the circumstances that your palette has to go above 7 it is good to group different factors you are trying to show and have similar colors with in that group.
Showing data is important but sometimes knowing how to show no data is as important as showing data. When we talk about showing no data usually we are talking about the back ground of your visual information. According to how you portrayed you data, different back ground colors can help or ruin your whole work. The background color has to disassociate itself from the any other colors on your graphic to be effective.