Font Format History

The Bitmap Era: 1960s-1980s

The earliest digital fonts were bitmap fonts, grids of pixels where each character was defined at a specific size. Each character existed as a rectangular array of on/off values. The representation was simple, fast to render, and perfectly suited to the raster displays and dot-matrix printers of the era. It was also fundamentally limited: a font defined at 12 points could only display at 12 points without visible degradation.

The concept dates to the 1960s, when mainframes used fixed-width bitmap characters for terminal output. Early displays like the IBM 3270 (1971) and DEC VT100 (1978) used character ROM chips containing bitmap glyphs. The Xerox Alto (1973), the first computer with a graphical user interface, used proportional bitmap fonts for on-screen display , an innovation that directly influenced the Macintosh interface a decade later.

Donald Knuth's Metafont system (1979) was among the earliest attempts at programmatically described scalable fonts, using mathematical equations to generate bitmaps at specified resolutions, an important intellectual predecessor to outline font technology.

Apple's original Macintosh (January 1984) shipped with bitmap fonts designed by Susan Kare, a graphic designer with no prior experience in digital typography. Kare created the original Mac system fonts, Chicago, Geneva, Monaco, and Cairo, as pixel art, carefully crafting each glyph at the specific sizes that would appear on the Mac's 72 dpi screen. Chicago remained the Mac's system font until 1999.

The fundamental limitation of bitmap fonts, that each size required a separate, hand-crafted bitmap file, was becoming untenable by the mid-1980s. A complete font family might require dozens of separate files for different sizes, weights, and styles. The industry needed a way to describe font shapes mathematically, so that a single description could be rendered crisply at any size.

The PostScript Revolution: 1984

Before PostScript, professional typesetting was the exclusive domain of specialists using dedicated, expensive equipment. A skilled typographer operating a Linotype or Phototypesetter machine was required to produce the kind of polished, proportionally spaced text readers expected in print. Desktop computers could produce text, but it was coarse bitmap-based output unsuitable for professional production.

Adobe Systems was founded in December 1982 by John Warnock and Charles Geschke, two researchers who had left Xerox PARC after Xerox declined to commercialize their work on a page description language called Interpress. Adobe released PostScript in 1984 as a device-independent page description language: a PostScript file describes a page in mathematical terms that can be rendered on any compatible device at any resolution.

The same file used to proof a document on a desktop laser printer could go directly to a high-end print service for final production without modification. This resolution independence was revolutionary. Adobe's first major customer was Apple Computer, Steve Jobs personally negotiated the deal that put PostScript inside the Apple LaserWriter printer, reportedly paying Adobe $2.5 million for a license.

Adobe Type 1 Fonts

Adobe Type 1 fonts, introduced alongside PostScript in 1984, were the first widely-adopted scalable outline fonts for personal computers. Type 1 used cubic Bezier curves (defined by four control points) to describe each glyph outline. Cubic curves are highly expressive, a single cubic segment can describe a wider range of curve shapes than a quadratic counterpart.

The format included "hints", instructions that improved rendering at small sizes by snapping outline features to pixel boundaries. Critically, Adobe encrypted this hinting data using a proprietary algorithm. While anyone could theoretically create PostScript fonts, only fonts using Adobe's hinting technology achieved optimal quality at text sizes on screen displays. This encryption was the foundation of Adobe's monopoly.

The LaserWriter and Desktop Publishing

The Apple LaserWriter, introduced in March 1985 at $6,995 (approximately $20,000 in 2026 dollars), was the first laser printer to include a built-in PostScript interpreter. The LaserWriter contained its own Motorola 68000 processor, as powerful as the Macintosh itself , and 512 KB of RAM dedicated entirely to running the PostScript interpreter and rendering pages.

Combined with Aldus PageMaker (released July 1985, developed by Paul Brainerd) and the Macintosh, the LaserWriter launched what Brainerd himself coined as "desktop publishing." This trio of technologies allowed small businesses, nonprofits, schools, and individual entrepreneurs to produce professional-quality printed materials for the first time without access to expensive commercial typesetting equipment.

The LaserWriter shipped with 13 built-in Type 1 fonts in ROM: four faces each of Times, Helvetica, and Courier, plus the Symbol typeface. These thirteen fonts became the foundation of a generation's typographic vocabulary. By 1987, the desktop publishing industry was worth over $1 billion. Adobe's PostScript licensing revenue grew from $1.6 million in 1984 to $16 million in 1986.

The Type 1 Monopoly

From 1984 to 1990, Adobe maintained a highly profitable monopoly on high-quality scalable font technology. The encrypted Type 1 specification was not public, forcing font foundries to license Adobe's technology or produce inferior results. Adobe controlled roughly 70% of the PostScript font market.

Adobe Type Manager (ATM), released in 1989, extended Type 1's reach by allowing PostScript fonts to render smoothly on screen without bitmap screen fonts. Previously, Type 1 fonts looked excellent when printed but appeared as jagged bitmaps on screen at sizes between the stored bitmap sizes. ATM used the PostScript outline data to render smooth glyphs at any screen size, dramatically improving on-screen experience.

The monopoly frustrated both Apple and Microsoft, who paid significant royalties to Adobe for font rendering on their platforms. Adobe's font business was among the most profitable segments of any technology company in the late 1980s. This dynamic would not last.

History of TrueType: 1989-1992

When Apple announced TrueType in 1989, it fundamentally altered the balance of power in digital type and ended Adobe's stranglehold on high-quality scalable fonts. Apple began development in the late 1980s as a direct response to Adobe's control over the PostScript font market. Apple's motivation was to create a royalty-free alternative that would give them control over font rendering on the Macintosh without depending on Adobe.

The project was initially codenamed "Bass" (later "Royal") within Apple, with access limited to a small team. TrueType was publicly announced at the Seybold publishing conference in September 1989 by Apple and Microsoft jointly. Apple CEO John Sculley presented the technology as a way to free the industry from Adobe's pricing control. The announcement forced Adobe to respond quickly, ultimately leading them to open the previously secret Type 1 specification in March 1990.

Sampo Kaasila and the Technical Design

Finnish-born engineer Sampo Kaasila was the primary architect of TrueType at Apple. He designed the TrueType rasterizer, the mathematical engine that converts outline font data into the pixel grids that screens display. Kaasila's rasterizer had to produce sharp, readable text at sizes as small as 8 or 9 pixels tall on the low-resolution screens of the late 1980s.

One of Kaasila's most significant contributions was the design of TrueType's hinting instruction set. TrueType hinting instructions are embedded as a bytecode program executed by the rasterizer, a virtual machine approach more powerful than PostScript Type 1's declarative hints. TrueType hinting allowed font designers to precisely control pixel placement at small sizes, snapping specific points to grid positions and controlling how strokes thickened or thinned at different resolutions.

The complexity of TrueType hinting became both its greatest strength and a practical limitation. While the system could produce exceptional results, creating high-quality TrueType hints required specialized expertise that few type designers possessed.

Quadratic vs Cubic Bezier Curves

TrueType's key technical decision was using quadratic Bezier curves rather than the cubic curves used by PostScript Type 1. This difference has practical implications for designers and tools that persist today.

At the time of TrueType's development (1987-1989), personal computers had limited processing power, and font rasterization needed to be fast on hardware without dedicated graphics processors. Quadratic Bezier curves require fewer floating-point operations than cubic curves. This is why OpenType, introduced in 1996, supports both outline types within the same container, fonts originally designed as TrueType retain quadratic outlines, while fonts designed in PostScript tools retain cubic CFF outlines.

Microsoft Licenses TrueType

In 1989, Microsoft licensed TrueType from Apple as part of a broader technology exchange agreement. Microsoft received TrueType technology, while Apple received access to Microsoft's virtual memory technology for use in future Mac operating systems. Microsoft engineer Greg Hitchcock led the effort to implement TrueType support in Windows.

Microsoft also invested heavily in creating TrueType versions of core fonts, commissioning the "Core fonts for the Web" project that produced Arial (a Helvetica-inspired design by Monotype), Times New Roman (Monotype's adaptation of the classic Times typeface), Courier New, Verdana (designed by Matthew Carter specifically for screen legibility), Georgia (also by Carter), and other typefaces that became de facto standards on Windows worldwide.

TrueType on Windows 3.1

Windows 3.1, released in April 1992, was the first Windows version with built-in TrueType support. Windows 3.1 shipped with 14 TrueType fonts including Arial, Times New Roman, Courier New, Symbol, and Wingdings. For the first time, every Windows user had access to scalable, high-quality fonts without purchasing additional software.

The impact was immediate. By 1993, TrueType had become the dominant font format on personal computers, installed on tens of millions of Windows machines worldwide. The .ttf file extension remains one of the most widely recognized font formats globally, TrueType outlines are still used in countless system fonts on Windows, Android, and many Linux distributions, and WOFF2 frequently contains TrueType outlines wrapped in compressed containers.

The Font Wars: Apple vs Adobe (1987-1996)

The history of modern font formats is, in large part, the story of a business conflict. Between 1984 and 1996, Adobe, Apple, and Microsoft fought over control of the fundamental infrastructure of digital typography. The fight was about money, licensing fees, market share, printer control, but the collateral damage was innovation: both sides were forced to build genuinely superior technology in their effort to win.

The Apple-Microsoft Alliance

Apple's TrueType announcement would have been far less significant if it had remained a Mac-only technology. Adobe could have weathered a format war on the Mac alone, the Windows market was far larger. Apple negotiated a licensing deal with Microsoft before the Seybold announcement.

In one of the more strategically improbable partnerships in technology history, Apple and Microsoft, fierce competitors in the operating system market, joined forces against Adobe. Apple licensed TrueType to Microsoft in exchange for Microsoft's virtual memory technology, which Apple needed for System 7. Both companies received something valuable; both shared an interest in breaking Adobe's font technology monopoly.

Adobe's Counter-Moves

Adobe was caught off guard by the TrueType announcement. The most significant concession came first: in March 1990, within six months of the TrueType announcement, Adobe published the complete Type 1 font specification in a book titled "Adobe Type 1 Font Format." This was a dramatic reversal of years of secrecy. The specification had been one of Adobe's most closely guarded assets, the technical foundation of its licensing business. Publishing it was an admission that the monopoly was unsustainable.

Adobe also released Adobe Type Manager (ATM) as a free download for both Mac and Windows, solving the problem of jagged Type 1 fonts on screen. ATM made Type 1 fonts render smoothly on screen at any size, matching a key advantage of TrueType.

In 1991, Adobe introduced Multiple Master fonts, an advanced Type 1 extension that allowed continuous variation between design extremes along multiple axes. Multiple Master could interpolate between light and heavy, condensed and extended, generating any intermediate instance on demand. This was technically impressive, a direct precursor to variable fonts , but never achieved mainstream adoption due to tooling complexity.

Despite these responses, TrueType's advantages were difficult to overcome. Its inclusion in both Mac OS and Windows, at no additional cost, gave it an insurmountable distribution advantage for the consumer and prosumer markets. By 1993-1994, TrueType was the dominant font format for personal computing. Type 1 retained its position in professional print publishing workflows, where existing investments in Type 1 font libraries created switching costs that TrueType couldn't overcome.

Legacy of the Font Wars

OpenType Unites the Industry: 1996

By the mid-1990s, the Font Wars had produced a fragmented market. Professional print workflows depended on Type 1; consumer and office computing ran on TrueType. Application developers maintaining design software had to support both formats. Font foundries had to create and maintain two versions of every typeface. The fragmentation created friction throughout the industry without producing any clear winner.

In 1994, Microsoft and Adobe began working together on a specification that would unify the two format families. OpenType, jointly announced in 1996, used TrueType's sfnt container and table structure. Within that container, OpenType supported both TrueType quadratic outlines and PostScript CFF (Compact Font Format) cubic outlines, allowing font designers to use either technology without consumers needing to care which was present.

More importantly, OpenType added the GSUB (Glyph Substitution) and GPOS (Glyph Positioning) tables, a standardized mechanism for advanced typographic features. Standard ligatures, discretionary ligatures, small capitals, old-style numerals, stylistic alternates, fractions, contextual alternates, swashes, historical forms, and dozens of other OpenType features became accessible through a standard interface accessible today via CSS font-feature-settings.

OpenType also addressed Unicode support comprehensively. A single OpenType font file could contain up to 65,535 glyphs and cover multiple writing systems, Latin, Cyrillic, Greek, Arabic, Hebrew, Indic scripts, CJK, and more, in one unified file. This laid the groundwork for the multilingual web typography of today.

Apple's relationship with OpenType was initially complex. Apple had invested in its own extension to TrueType called AAT (Apple Advanced Typography), which provided similar advanced features through a different mechanism. Apple eventually supported OpenType in Mac OS X while maintaining AAT support, resulting in a period where font designers needed to encode advanced features in both GSUB/GPOS tables (for OpenType) and morx/kern tables (for AAT). This gradually resolved as OpenType achieved universal adoption.

OpenType received ISO standardization in 2007 as ISO/IEC 14496-22, formalizing its status as an international standard. As of 2026, OpenType (in its various versions through 1.9) remains the universal desktop font format, with over 99% of professionally produced fonts available in OpenType.

Web Fonts Evolution: 1997-2014

For the first decade of the web, designers worked within the prison of "web-safe fonts", a handful of typefaces installed on virtually every computer that could be safely referenced in CSS. Everything outside that small set was a gamble: you could specify Garamond or Futura, but users without those fonts installed would see Times New Roman or Arial instead.

CSS2 @font-face and EOT

The CSS2 specification, published by W3C in May 1998, included the @font-face rule, a mechanism for web pages to reference external font files. In practice, implementation was inconsistent and the feature effectively disappeared from practical use within a few years of introduction.

Internet Explorer 4 (1997) and IE 5 (1999) supported @font-face, but exclusively with Microsoft's proprietary EOT (Embedded OpenType) format. EOT was a compressed, DRM-protected version of TrueType or OpenType fonts designed to prevent font piracy. EOT files bound fonts to specific domains using a digital rights management mechanism. Microsoft created the WEFT (Web Embedding Fonts Tool) to generate EOT files. The process was cumbersome and inaccessible. No other browser supported EOT. Microsoft submitted EOT to the W3C in 2007-2008, but it was rejected in favor of developing WOFF.

The Dark Ages: sIFR and Cufon

Between 2004 and 2009, web designers invented elaborate workarounds. sIFR (Scalable Inman Flash Replacement, 2004) used Adobe Flash to render custom text, a JavaScript file scanned a page for headings, hid the HTML text, and injected Flash movies rendering the same text using a custom font. Cufon (2009) avoided Flash by pre-converting font outlines into JavaScript, then rendering text onto HTML5 canvas elements.

Typekit and the Font Service Revolution

In September 2009, Jeffrey Veen, Bryan Mason, Ryan Carver, and Greg Veen launched Typekit , the first practical commercial web font hosting service. Type foundries licensed fonts to Typekit; Typekit served them via JavaScript and CSS to paying subscribers; the technical complexity of cross-browser compatibility was Typekit's problem.

Typekit demonstrated massive pent-up demand for quality web fonts. Within months it secured partnerships with major foundries including Font Bureau, Dalton Maag, and Mark Simonson Studio. Adobe acquired Typekit in October 2011 for approximately $30 million and eventually integrated it into Creative Cloud as Adobe Fonts.

Google Fonts Democratizes Web Typography

Google Fonts launched in May 2010, offering a completely free, open-source font library served from Google's global CDN. While Typekit had proven web fonts were desirable and commercially viable, Google Fonts removed every remaining barrier: no cost, no licensing complexity, no subscription, no JavaScript dependency. Key commissions included Roboto (Christian Robertson, 2011, Android's default), Open Sans (Steve Matteson, 2011), Lato (Lukasz Dziedzic, 2010), and the Noto project covering every Unicode character.

WOFF: The First True Web Font Format

WOFF (Web Open Font Format) was proposed in 2009 by Erik van Blokland, Tal Leming, and Mozilla engineer Jonathan Kew. WOFF wasn't a new outline format, it contained existing TrueType or OpenType data unchanged, but a packaging format specifically designed for web delivery. WOFF wrapped existing font tables in a compressed container, reducing file sizes by 40-50% compared to uncompressed TrueType, and added standardized metadata fields.

Why WOFF2 Was Created (2014)

WOFF was good, but as mobile internet usage exploded and Google began deep research into web performance, the Google Fonts team identified web fonts as one of the most significant bottlenecks affecting page load time worldwide. A purpose-built compression scheme exploiting the internal structure of font data could do considerably better than zlib.

The Limitations of WOFF

WOFF 1.0's fundamental limitation was that it applied zlib to raw font table data without preprocessing. Font data has highly predictable internal structures, glyph coordinates follow geometric patterns, metrics tables have regular shapes, character mapping data is largely sequential, but zlib was never designed to exploit these domain-specific patterns. A compressor that understood font internals could do significantly better.

Google's research showed font files were among the largest render-blocking resources on many web pages. The median web page loaded 3-4 font files totaling 100-300 KB compressed. On 3G mobile connections common in developing markets during 2012-2015, this could add 1-3 seconds to first contentful paint.

Google's Brotli Innovation

Brotli is a general-purpose compression algorithm developed by Google engineers Jyrki Alakuijala and Zoltán Szabadka, initially released in 2013 specifically for WOFF2. Brotli was later generalized for HTTP content encoding and accepted by all major browsers between 2016 and 2017.

The WOFF2 Glyph Transform

The secret weapon of WOFF2 is its glyph transform preprocessing step. Before Brotli compression is applied, WOFF2 restructures the TrueType glyph data (the 'glyf' table) to be more compressor-friendly. This transformation accounts for roughly half of WOFF2's improvement over WOFF 1.0.

In a standard TrueType font, each glyph's outline data interleaves flag bytes, x-coordinates, and y-coordinates, terrible for compression because these have very different statistical distributions. The WOFF2 glyph transform separates glyph data into parallel streams: one stream for all flags, one for all x-coordinates, one for all y-coordinates, and one for composite glyph information. The transform also applies delta encoding , storing the difference between consecutive point positions, producing small numbers clustered near zero that compress exceptionally well.

Compression Comparison: WOFF vs WOFF2

Standardization and Browser Adoption

WOFF2 in Practice Today

Variable Fonts and Beyond: 2016-Present

OpenType 1.8, published in September 2016, introduced font variations, commonly known as variable fonts. The specification was developed jointly by Apple, Google, Microsoft, and Adobe, representing the most significant cooperative effort in font technology since OpenType itself. Variable fonts allow a single font file to contain a continuous design space with multiple variation axes, rather than discrete named instances.

The standard axes defined in OpenType 1.8 include weight (wght), width (wdth), italic (ital), slant (slnt), and optical size (opsz). Font designers can also define custom axes with four-character tags. Registered axes are accessed via familiar CSS properties like font-weight and font-stretch; custom axes use font-variation-settings.

The performance benefits of variable fonts for web use are significant. A type family previously requiring eight separate weight files (Thin through Black) can be replaced by a single variable font file that is often smaller than just two of the static files it replaces. Browser support achieved broad coverage by 2018: Chrome 62 (October 2017), Firefox 62 (September 2018), Safari 11 (September 2017), and Edge 17 (April 2018).

Apple introduced a conceptual precursor to variable fonts in 1991 with TrueType GX Variations, which offered a similar design space concept but never achieved wide adoption due to limited tool support. The OpenType variable font specification incorporated and refined these concepts in a fully cross-platform implementation.

The current frontier of font technology includes per-glyph subsetting (serving only the exact characters used on each page, computed at request time using the unicode-range descriptor), and continued improvements to variable font tooling. The next major OpenType version is expected to formalize features currently in development.

The Complete Timeline

Every major font format milestone from the 1960s to the present, in one chronological reference.