We then used iWantHue to expand the light color palette to a total of 10 colors (as shown in the image below). The result was seven “light” colors and five “dark” colors. That way, we wouldn’t arbitrarily place a very light color with two dark colors and cause it to stand out or create other biases in our data visualizations. With the Redwood colors plugged in to iWantHue, our first order of business was to separate out the colors into light and dark. We’re always working in the land of tradeoffs and compromises. Color perception is always a relative and somewhat individual affair. (Check out the tutorials and theory sections on iWantHue or Google “CIE Lab color space” to learn more about color and visual perception.) Also remember that the perception of colors changes dramatically in different circumstances. Generally speaking, the narrower the range on “C” and “L”, the more likely an undetermined set of colors is to be balanced. When we use iWantHue to design custom palettes for the purpose of data visualization, we try to limit the range of both the “C” and “L” values so that we get a balanced palette in terms of visual weight. We categorized the colors either as “dark” or “light” based on the “L” value and based on their shape. ![]() Notice that the dark green color has a mostly triangular shape to its projection of the three-dimensional saddle onto the screen while the lighter orange color has a generally rectangular (rhombus) shape to it. Below is a screenshot of the first two colors. The developers that built iWantHue have included tools that help us understand how the human eye perceives colors, both individually and in pair-wise comparisons. Furthermore, some hues appear to be brighter or have more presence than others. They may be close in one dimension but further away in others) Our eyes naturally distinguish between certain colors more than others and we can see more range for some hues than others. (A color space is a mathematical abstraction of the position of different colors relative to all the others. The CIELAB color space shown here is a three-dimensional “twisted saddle” shape. The screenshot below shows the full color space available and the first 10 colors of the Redwood data visualization palette.Īs we edit each individual color, we get a representation of the color space that it belongs to. To do that, we want to use colors of similar visual weight and intensity so that we don’t unintentionally call more attention to some members of a categorical set than others. Our general goal for data visualization is to represent data values in a way that is as objective and non-biased as possible. iWantHue automatically generates color palettes of optimally distinct colors by letting users refine the Hue (H), Chroma (C), and Lightness (L) of a three-dimensional array of colors. The iWantHue site was built specifically for use by data scientists and has a tremendous amount of information about colors and how they appear in human perception. We wanted to learn more about these colors so we plugged them into a website that we often use to create balanced color palettes called iwanthue. These 12 colors were put into Oracle Analytics in the order shown below.Īnalyzing the Default Data Visualization Colors The Oracle JET JavaScript Extension Toolkit for Redwood specification (which you can read here: ) identifies 12 colors for data visualization. While it’s clear that the Oracle design team has thought a lot about user interface design principles in general, we have some recommendations for using Redwood that should help some of your data visualizations in Oracle Analytics Cloud. In the world of visual analytics, the strategy behind color choices for various graph elements is extremely important. ![]() In this blog, however, we are going to focus on the use of color in visualizations. Redwood introduces a unified look and feel across all Oracle products that includes fonts, color palettes, menu layouts, and a consistent strategy behind callouts. The introduction of the Redwood theme in Oracle Analytics represents the culmination of Oracle’s long-term, serious commitment to leading practices in user interface design.
0 Comments
Leave a Reply. |