ABSTRACT
In data visualizations, point marks can be replaced with labels for measurably faster identification than cross-referencing a legend or performing an interaction, such as with a tooltip. Location and proximity tasks may be faster as well. Alphanumeric codes can be used for consistent label sizes or variable length labels can be anchored to visually dominant point marks. With many points, aggregate labels can extend across groups of points, short codes substituted, or label visibility can be yoked to interaction, for example, with more labels made visible on zoom. Examples include scatterplots, a bubble plot, a ternary plot, and graphs showing communities extracted from categorizations and social media.
