What Is a Typeface? Design That Works Best When Nobody Notices It

Typography has a precise vocabulary because the differences it describes are small and consequential:

• •The x-height, the height of a lowercase x, which affects apparent size far more than the nominal point size does, so two faces set at the same size can look markedly different
• •Ascenders and descenders, the parts rising above the x-height and dropping below the baseline, whose length affects how tightly lines can be set
• •Serifs, the small strokes finishing a letter, which distinguish serif faces from sans serif ones and come in bracketed, slab and hairline forms
• •Stroke contrast, the difference between thick and thin parts, which is high in faces descending from pointed pen writing and low in those built from a broad even stroke
• •Counters, the enclosed spaces inside letters like o and e, whose openness affects legibility at small sizes and in poor conditions
• •The aperture, how open the gap is in letters like c, s and a, which strongly affects how easily letters are told apart
• •Kerning and tracking, the adjustment of space between specific letter pairs and across a whole run, which are separate operations and frequently confused

The conventional historical grouping is useful because each category came from a different set of tools and purposes. Humanist or old style faces, from the fifteenth and sixteenth centuries, derive from writing with a broad nib held at an angle, so the stress in a round letter is diagonal and the contrast is moderate. Transitional faces of the eighteenth century straightened that stress and increased contrast. Didone or modern faces pushed contrast to an extreme with hairline serifs, made possible by smoother paper and better presses, and are elegant and fragile at small sizes. Slab serifs arrived in the nineteenth century for advertising, where weight mattered more than refinement. Sans serif faces, initially called grotesques, were also nineteenth-century commercial letterforms, later reinterpreted by modernists into the geometric faces built from circles and straight lines and the neutral neo-grotesques of the mid twentieth century. Script, display and blackletter categories cover everything designed for a specific effect rather than for continuous reading.

The two words are used interchangeably and should not be. Legibility is a property of the letterforms, meaning how easily individual characters are distinguished, which matters most for road signs, labels and anything read under pressure and which is why signage faces open their apertures and differentiate confusable pairs like capital I, lowercase l and the digit one. Readability is a property of the setting, meaning how comfortably continuous text can be read, and depends on size, line length, line spacing, contrast and word spacing at least as much as on the face. The common belief that serif faces are more readable than sans serif ones in print has not survived testing, with studies repeatedly finding no reliable difference once other variables are controlled, and the practical rule is that familiarity dominates, since readers read best what they read most. Line length matters demonstrably, with something in the region of sixty to seventy-five characters per line usually recommended, since very long lines make it hard to find the start of the next one.

Type designed for metal and paper behaved badly on early screens, where a letter had to be built from a coarse grid of pixels and fine detail simply disappeared. The response was faces designed for the constraint, with hinting instructions telling the renderer how to snap outlines to the pixel grid, and generous x-heights and open shapes to survive low resolution. High-density displays removed most of that pressure, and the discipline shifted to different problems: how a face performs across many sizes and weights on unknown devices, how to load type over a network without leaving a page blank or flashing a substitution, and how to support the enormous character sets that genuine multilingual typography requires. Variable fonts, introduced in 2016, changed the technical model by storing a single outline with axes of variation, so weight, width and optical size can be adjusted continuously from one file rather than shipping dozens, which is both an efficiency gain and a return to something type founders always understood, that a face wants to be drawn differently at different sizes.

A typeface is a designed set of letterforms and a font is one instance of it, and the craft succeeds when the reader notices nothing. X-height, stroke contrast, counters and aperture matter more to how type performs than the name of the face. Legibility concerns telling characters apart and readability concerns reading comfortably at length, which is governed mainly by size, line length and spacing. The claim that serif faces read better than sans serif ones has not held up under testing.