Mapping is the process of applying encodings to data (called [[marks]] in the context of [[data visualization]]). Marks can be represented using position, color, shape, orientation, size, opacity, motion, and texture. Mapping should reflect human perception, an area of ongoing research in data visualization. Mapping should be **expressive** and **effective**. Expressiveness states that the visual encoding should express all of, and only, the information in the dataset attributes. For example, use opacity/lightness of color to represent [[ordinal]] data and different hues to represent [[categorical]] data. Effectiveness states that the importance of attribute should match the salience of the channel. In other words, a larger mark should reflect a larger (or more important) value.