Typeface

The size of typefaces and fonts is traditionally measured in points;[5] point has been defined differently at different times, but now the most popular is the Desktop Publishing point of 1⁄72 in (0.0139 in or 0.35 mm). When specified in typographic sizes (points, kyus), the height of an em-square, an invisible box which is typically a bit larger than the distance from the tallest ascender to the lowest descender, is scaled to equal the specified size.[6] For example, when setting Helvetica at 12 point, the em square defined in the Helvetica font is scaled to 12 points or 1⁄6 in or 4.2 mm. Yet no particular element of 12-point Helvetica need measure exactly 12 points.

Frequently measurement in non-typographic units (feet, inches, meters) will be of the cap-height, the height of the capital letters. Font size is also commonly measured in millimeters (mm) and qs (a quarter of a millimeter, kyu in romanized Japanese) and inches.

	Sans serif font
	Serif font
	Serif font with serifs highlighted in red

Typefaces can be divided into two main categories: serif and sans serif. Serifs comprise the small features at the end of strokes within letters. The printing industry refers to typeface without serifs as sans serif (from French sans, meaning without), or as grotesque (or, in German, grotesk).

Great variety exists among both serif and sans serif typefaces. Both groups contain faces designed for setting large amounts of body text, and others intended primarily as decorative. The presence or absence of serifs represents only one of many factors to consider when choosing a typeface.

Typefaces with serifs are often considered easier to read in long passages than those without. Studies on the matter are ambiguous, suggesting that most of this effect is due to the greater familiarity of serif typefaces. As a general rule, printed works such as newspapers and books almost always use serif typefaces, at least for the text body. Web sites do not have to specify a font and can simply respect the browser settings of the user. But of those web sites that do specify a font, most use modern sans serif fonts, because it is commonly believed that, in contrast to the case for printed material, sans serif fonts are easier than serif fonts to read on the low-resolution computer screen.

A proportional typeface contains glyphs of varying widths, while a monospaced (non-proportional or fixed-width) typeface uses a single standard width for all glyphs in the font. Duospaced fonts are similar to monospaced fonts, but characters can also be two character widths instead of a single character width.

Many people generally find proportional typefaces nicer-looking and easier to read, and thus they appear more commonly in professionally published printed material. For the same reason, GUI computer applications (such as word processors and web browsers) typically use proportional fonts. However, many proportional fonts contain fixed-width (tabular) figures so that columns of numbers stay aligned.

Monospaced typefaces function better for some purposes because their glyphs line up in neat, regular columns. No glyph is given any more weight than another. Most manually operated typewriters use monospaced fonts. So do text-only computer displays and third- and fourth-generation game console graphics processors, which treat the screen as a uniform grid of character cells. Most computer programs which have a text-based interface (terminal emulators, for example) use only monospaced fonts (or add additional spacing to proportional fonts to fit them in monospaced cells) in their configuration. Monospaced fonts are commonly used by computer programmers for displaying and editing source code so that certain characters (for example parentheses used to group arithmetic expressions) are easy to see.[9]

ASCII art usually requires a monospaced font for proper viewing, with the exception of Shift JIS art which takes advantage of the proportional characters in the MS PGothic font. In a web page, the <tt> </tt>, <code> </code> or <pre> </pre> HTML tags most commonly specify monospaced fonts. In LaTeX, the verbatim environment or the Teletype font family (e.g., \texttt{...} or {\ttfamily ...}) uses monospaced fonts (in TeX, use {\tt ...}).

Any two lines of text with the same number of characters in each line in a monospaced typeface should display as equal in width, while the same two lines in a proportional typeface may have radically different widths. This occurs because in a proportional font, glyph widths vary, such that wider glyphs (typically those for characters such as W, Q, Z, M, D, O, H, and U) use more space, and narrower glyphs (such as those for the characters i, t, l, and 1) use less space than the average.

In the publishing industry, it was once the case that editors read manuscripts in monospaced fonts (typically Courier) for ease of editing and word count estimates, and it was considered discourteous to submit a manuscript in a proportional font. This has become less universal in recent years, such that authors need to check with editors as to their preference, though monospaced fonts are still the norm.

The word Sphinx, set in Adobe Garamond Pro to illustrate the concepts of baseline, x-height, body size, descent and ascent.

Most scripts share the notion of a baseline: an imaginary horizontal line on which characters rest. In some scripts, parts of glyphs lie below the baseline. The descent spans the distance between the baseline and the lowest descending glyph in a typeface, and the part of a glyph that descends below the baseline has the name descender. Conversely, the ascent spans the distance between the baseline and the top of the glyph that reaches farthest from the baseline. The ascent and descent may or may not include distance added by accents or diacritical marks.

In the Latin, Greek and Cyrillic (sometimes collectively referred to as LGC) scripts, one can refer to the distance from the baseline to the top of regular lowercase glyphs (mean line) as the x-height, and the part of a glyph rising above the x-height as the ascender. The distance from the baseline to the top of the ascent or a regular uppercase glyphs (cap line) is also known as the cap height.[10] The height of the ascender can have a dramatic effect on the readability and appearance of a font. The ratio between the x-height and the ascent or cap height often serves to characterize typefaces.

Typefaces that can be substituted for one another in a document without changing the document's text flow are said to be "metrically identical" (or "metrically compatible").[11][12][13][14][15] Several typefaces have been created to be metrically compatible with widely used proprietary typefaces to allow the editing of documents set in such typefaces in digital typesetting environments where these typefaces are not available. For instance, the free and open-source Liberation fonts and Croscore fonts have been designed as metrically compatible substitutes for widely used Microsoft fonts.[16][17][18]

During the metal type era, all type was cut in metal and could only be printed at a specific size. It was a natural process to vary a design at different sizes, making it chunkier and clearer to read at smaller sizes.[19][20] Many digital typefaces are offered with a range of fonts (or a variable font axis) for different sizes, especially designs sold for professional design use. The art of designing fonts for a specific size is known as optical sizing. Others will be offered in only one style, but optimised for a specific size. Optical sizes are particularly common for serif fonts, since the fine detail of serif fonts can need to be bulked up for smaller sizes.[21][22][23]

Typefaces may also be designed differently considering the type of paper on which they will be printed. Designs to be printed on absorbent newsprint paper will be more slender as the ink will naturally spread out as it absorbs into the paper, and may feature ink traps: areas left blank into which the ink will soak as it dries. These corrections will not be needed for printing on high-gloss cardboard or display on-screen. Fonts designed for low-resolution displays, meanwhile, may avoid pure circles, fine lines and details a screen cannot render.[24]

Hoefler Text uses non-lining or lower-case figures.

Proportional (left-side) and tabular (right-side) numeric digits, drawn as lining figures.

Most typefaces, especially modern designs, include a complementary set of numeric digits.[25]

Numbers can be typeset in two main independent sets of ways: lining and non-lining figures, and proportional and tabular styles.[lower-alpha 2]

Most modern typefaces set numeric digits by default as lining figures, which are the height of upper-case letters. Non-lining figures, styled to match lower-case letters, are often common in fonts intended for body text, as they are thought to be less disruptive to the style of running text. They are also called lower-case numbers or text figures for the same reason.

The horizontal spacing of digits can also be proportional, with a character width tightly matching the width of the figure itself, or tabular, where all digits have the same width. Proportional spacing places the digits closely together, reducing empty space in a document, and is thought to allow the numbers to blend into the text more effectively.[26] As tabular spacing makes all numbers with the same number of digits the same width, it is used for typesetting documents such as price lists, stock listings and sums in mathematics textbooks, all of which require columns of numeric figures to line up on top of each other for easier comparison.[27] Tabular spacing is also a common feature of simple printing devices such as cash registers and date-stamps.[28]

Characters of uniform width are a standard feature of so-called monospaced fonts, used in programming and on typewriters. However, many fonts that are not monospaced use tabular figures. More complex font designs may include two or more combinations with one as the default and others as alternate characters.[29] Of the four possibilities, non-lining tabular figures are particularly rare since there is no common use for them.[30][31][32]

Fonts intended for professional use in documents such as business reports may also make the bold-style tabular figures take up the same width as the regular (non-bold) numbers, so a bold-style total would appear just as wide as the same sum in regular style.[33][34][35]

The three traditional styles of serif typefaces used for body text: old-style, transitional and Didone, represented by Garamond, Baskerville and Didot.

Serif, or Roman, typefaces are named for the features at the ends of their strokes. Times New Roman and Garamond are common examples of serif typefaces. Serif fonts are probably the most used class in printed materials, including most books, newspapers and magazines. Serif fonts are often classified into three subcategories: Old Style, Transitional, and Didone (or Modern), representative examples of which are Garamond, Baskerville, and Bodoni respectively.

Old Style typefaces are influenced by early Italian lettering design.[36] Modern fonts often exhibit a bracketed serif and a substantial difference in weight within the strokes. Though some argument exists as to whether Transitional fonts exist as a discrete category among serif fonts, Transitional fonts lie somewhere between Old Style and Modern style typefaces. Transitional fonts exhibit a marked increase in the variation of stroke weight and a more horizontal serif compared to Old Style. Slab serif designs have particularly large serifs, and date to the early nineteenth century. The earliest known slab serif font was first shown around 1817 by the English typefounder Vincent Figgins.[37]

Roman, italic, and oblique are also terms used to differentiate between upright and two possible slanted forms of a typeface. Italic and oblique fonts are similar (indeed, oblique fonts are often simply called italics) but there is strictly a difference: italic applies to fonts where the letter forms are redesigned, not just slanted. Almost all serif faces have italic forms; some sans-serif faces have oblique designs. (Most faces do not offer both as this is an artistic choice by the font designer about how the slanted form should look.)[38]

The sans-serif Helvetica (Neue Haas Grotesk) typeface

Sans serif (lit. without serif) designs appeared relatively recently in the history of type design. The first, similar to slab serif designs, was shown in 1816 by William Caslon IV. Many have minimal variation in stroke width, creating the impression of a minimal, simplified design. When first introduced, the faces were disparaged as "grotesque" (or "grotesk") and "gothic":[39] but by the late nineteenth century were commonly used for san-serif without negative implication.[40]

The major sub-classes of Sans-serif are "Grotesque", "Neo-grotesque", "Geometric" and "Humanist".

"Blackletter" is the name of the class of typefaces used with the earliest printing presses in Europe, which imitated the calligraphy style of that time and place. Various forms exist including textualis, rotunda, schwabacher and fraktur. (Some people refer to Blackletter as "gothic script" or "gothic font", though the term "Gothic" in typography refers to sans serif typefaces.[40])

Gaelic fonts were first used for the Irish language in 1571, and were used regularly for Irish until the early 1960s, though they continue to be used in display type and type for signage. Their use was effectively confined to Ireland, though Gaelic typefaces were designed and produced in France, Belgium, and Italy. Gaelic typefaces make use of insular letterforms, and early fonts made use of a variety of abbreviations deriving from the manuscript tradition.[41][42] Various forms exist, including manuscript, traditional, and modern styles, chiefly distinguished as having angular or uncial features.[43]

Courier, a monospaced slab serif typeface. All the letters occupy spaces the same width.

Monospaced fonts are typefaces in which every glyph is the same width (as opposed to variable-width fonts, where the w and m are wider than most letters, and the i is narrower). The first monospaced typefaces were designed for typewriters, which could only move the same distance forward with each letter typed. Their use continued with early computers, which could only display a single font. Although modern computers can display any desired typeface, monospaced fonts are still important for computer programming, terminal emulation, and for laying out tabulated data in plain text documents; they may also be particularly legible at small sizes due to all characters being quite wide.[44] Examples of monospaced typefaces are Courier, Prestige Elite, Fixedsys, and Monaco. Most monospaced fonts are sans-serif or slab-serif as these designs are easiest to read printed small or display on low-resolution screens, though many exceptions exist.

CJK, or Chinese, Japanese and Korean typefaces consist of large sets of glyphs. These typefaces originate in the glyphs found in brush calligraphy during the Tang dynasty. These later evolved into the Song style (宋体字) which used thick vertical strokes and thin horizontal strokes in wood block printing.[45]

The glyphs found in CJK fonts are designed to fit within a square. This allows for regular vertical, horizontal, right-to-left and left-to-right orientations. CJK fonts can also include an extended set of monospaced Latin characters. This commonly results in complex, sometimes contradictory rules and conventions for mixing languages in type.

With CJK typefaces, Mincho style tends to be something like Serifs for the end of stems, and in fact includes Serifed glyphs for Extended Latin and Cyrillic sets within a typeface.

With CJK typefaces, Goth style tends to be something like Sans Serifs with squarish, cut off end-caps for the end of stems, and in fact includes Sans Serif glyphs for Extended Latin and Cyrillic sets within a typeface.

With CJK typefaces, Maru style tends to be something like Sans Serifs with rounded end-caps for the end of stems, and in fact includes Rounded Sans Serif glyphs for Extended Latin and Cyrillic sets within a typeface.

Coronet, a script typeface

Script typefaces imitate handwriting or calligraphy. They do not lend themselves to quantities of body text, as people find them harder to read than many serif and sans-serif typefaces; they are typically used for logos or invitations. Historically, most lettering on logos, displays, shop frontages did not use fonts but was rather custom-designed by signpainters and engravers, so many emulate the styles of hand-drawn signs from different historical periods. The genre has developed rapidly in recent years due to modern font formats allowing more complex simulations of handwriting.[49] Examples include Coronet (a quite simple design from 1937) and Zapfino (a much more complicated digital design).

Simulated Hebrew

Mimicry typefaces are decorative typefaces that have been designed to represent characters of one alphabet but at the same time evoke another writing system.[50] This group includes Roman typefaces designed to appear as Arabic, Chinese characters (Wonton fonts), Cyrillic (Faux Cyrillic), Indic scripts, Greek (an example being Lithos), Hebrew (Faux Hebrew), Kana, or Thai. These are used largely for the purpose of novelty to make something appear foreign, or to make businesses offering foreign products, such as restaurants, clearly stand out.[51][52][53] This typographic mimicry is also known as a faux font (named faux x, where x is usually a language script), pseudoscript, ethnic typeface, simulation typeface or a "foreign look" font.[54][55][56]

Reverse-contrast type compared to a fat face design. Both are very bold, but the fat face's thick lines are the verticals and the reverse-contrast's are the horizontals.

A reverse-contrast type is a typeface in which the stress is reversed from the norm: instead of the vertical lines being the same width or thicker than horizontals, which is normal in Latin-alphabet printing, the horizontal lines are the thickest.[57] Reverse-contrast types are rarely used for body text, and are particularly common in display applications such as headings and posters, in which their unusual structure may be particularly eye-catching.[58] First seen in London in 1821, they were particularly common in the mid- to late nineteenth century in American and British printing and have been revived occasionally since then. They effectively become slab serif designs because of the serifs becoming thick, and are often characterised as part of that genre. In recent times, the reverse-contrast effect has been extended to other kinds of typeface, such as sans-serif designs.[59]

Three typefaces designed for headings, offering a clear contrast to body text

Some typefaces have a structure that suggests a three-dimensional letter, such as letters carved into stone. An example of this is the genre known as 'inline', 'block' 'outline' or 'shadowed' typefaces. This renders the interior of glyphs in the background color, with a thin line around the edges of the glyphs. In some cases, the outline shows the glyph filled in with the foreground color, surrounded a thin outline mirroring the edges separated by a small gap. (This latter style is often used with "college" typefaces.) Colorized block lettering is often seen in carefully rendered graffiti.

A "shadow" effect can also be either designed into a typeface or added to an existing typeface. Designed-in shadows can be stylized or connected to the foreground. An after-market shadow effect can be created by making two copies of each glyph, slightly offset in a diagonal direction and possibly in different colors. Drop shadows can also be dynamically created by rendering software. The shadow effect is often combined with the outline effect, where the top layer is shown in white with black outline and the bottom layer in black, for greater contrast. An example typeface with an 'inline' effect is Imprint Shadowed, where the shadowed version is more widely distributed than the regular design.[60]

Some typefaces are specifically designed to be printed at small sizes, for example in telephone directories or on newsprint paper. Bell Gothic and Bell Centennial, commissioned for telephone directories, are notable examples of this. Small-print designs often feature a large x-height, and a chunky design. Some fonts used at such sizes may be members of a larger typeface family joining members for normal sizes. For example, the Times New Roman family contains some designs intended for small print use, as do many families with optical sizes such as Minion.

In the metal type era, typefaces intended to be printed small contained ink traps, small indentations at the junctions of strokes that would be filled up with ink spreading out, maintaining the intended appearance of the type design. Without ink traps, the excess ink would blob and ruin the crisp edge. At larger sizes, these ink traps were not necessary, so display faces did not have them. They have also been removed from most digital fonts, as these will normally be viewed on screen or printed through inkjet printing, laser printing, offset lithography, electrophotographic printing or other processes that do not show the ink spread of letterpress. Ink traps have remained common on designs intended to be printed on low-quality, absorbent paper, especially newsprint and telephone directories.

Ornamental (also known as novelty or sometimes display) typefaces are used to decorate a page. Historically complex interlocking patterns known as arabesques were common in fine printing, as were floral borders known as fleurons evoking hand-drawn manuscripts.

In the metal type era, type-founding companies often would offer pre-formed illustrations as fonts showing objects and designs likely to be useful for printing and advertisements, the equivalent of modern clip art and stock photographs.[65] As examples, the American Type Founders specimen of 1897 offered designs including baseball players, animals, Christmas wreaths, designs for cheques, and emblems such as state seals for government printing.[66] The practice has declined as printing custom illustrations and colour printing using processes such as lithography has become cheaper, although illustration typefaces are still sold by some companies. See above for the historical definition of display typeface.

Examples of dingbats, which could be used in documents such as tourist guides or TV listings.

Symbol, or dingbat, typefaces consist of symbols (such as decorative bullets, clock faces, railroad timetable symbols, CD-index, or TV-channel enclosed numbers) rather than normal text characters. Common, widely used symbol typeface releases include Zapf Dingbats and Wingdings, though many may be created internally by a publication for its own use and some typefaces may have a symbol range included.[67] Marlett is an example of a font used by Windows to draw elements of windows and icons.

Emoji are pictograms that can be used and displayed inline with text.[68][69] They are similar to previous symbol typefaces, but with a much larger range of characters, such as symbols for common objects, animals, food types, weather and emotions. Originally developed in Japan, they are now commonly installed on many computer and smartphone operating systems.[70][71] Following standardisation and inclusion in the Unicode standard, allowing them to be used internationally, the number of Emoji characters has rapidly increased to meet the demands of an expanded range of cultures using them; unlike many previous symbol typefaces, they are interchangeable with the ability to display the pictures of the same meaning in a range of fonts on different operating systems.[72][73] The popularity of emoji has meant that characters have sometimes gained culture-specific meanings not inherent to the design.[74][75][76] Both colour and monochrome emoji typefaces exist, as well as at least one animated design.[77]

Typefaces that include musical notes and other needed symbols have been developed to print sheet music.