Mean Hand

Mean Hand is a typeface built from 814,255 handwritten characters collected by the U.S. government in the early 1990s to automate Census form processing. The samples came from Census Bureau employees and high school students in Bethesda, Maryland. This data became NIST Special Database 19, later adapted into the EMNIST dataset: a benchmark used to train and evaluate handwriting recognition systems, shaping what counts as legible.

Each letter is constructed by stacking thousands of samples and applying a threshold: a mark appears only if that proportion of samples placed ink there. This threshold determines the weight.

At Black, 1 in 20 samples is sufficient; letters accumulate nearly every variation, becoming dense and difficult to read. At Regular, 1 in 3 samples must agree, producing forms that are readable but belong to no individual hand. At Thin, 3 in 4 samples must align; very little survives, and some letters become hard to distinguish.

At both extremes the type is illegible: overwhelmed by variation at one end, reduced to fragments at the other. Legibility emerges only within a narrow threshold. But even there, what resembles handwriting was written by no one.

Because EMNIST defines the baseline for what counts as legible handwriting, its distribution shapes whose writing gets recognized and whose does not. Mean Hand is that distribution made visible as type, available for free download.