The Digital Past: When Typefaces Were Experimental
By Paul Shaw May 19, 2005
The Digital Past: When Typefaces Were Experimental
By Paul Shaw May 19, 2005
The Digital Past: When Typefaces Were Experimental
By Paul Shaw May 19, 2005

Voice asked me to do a post-digital exploration of type to see if a new “experimental” stage was in the wings. But the more I thought about his request, the more I felt it was necessary to look back at the history of digital type design and sort out what really happened before trying to look ahead.


In the early 1990s, the term “experimental” came to be associated with fonts displayed in Emigre magazine. A number of those fonts were worthy of the label, but in the ensuing years “experimental” was attached to any typeface that seemed to be outside the norm. It came to mean wild, radical or weird. In short, it became a marketing device rather than a useful description.

Instead of looking solely, or even primarily, at such “experimental” fonts I have chosen to list typefaces that for one reason or another—technological, ideological, conceptual, cultural or even aesthetic—broke new ground in the digital era. This is not a list of the most beautiful, coolest or most popular fonts. (Nor is it simply a list of my favorite fonts.) This list is a reminder that the digital era has not been static. Like the metal era, it has had its own stages of development, each of which has affected not only how type is made, but how it looks and how it is used.


Digi-Grotesk S

Despite its derivative appearance—it looks like a condensed knock-off of Neuzeit Book from Stempel—Digi-Grotesk S is significant because it was the first digital typeface. It was created by the staff of Dr. Ing. Rudolf Hell, the company that pioneered cathode ray phototypesetters.



Designed by Hermann Zapf for Hell, Marconi was the first original typeface to be produced with the Ikarus computer-aided design and digitization system. Ikarus, developed in 1973 by Peter Karow of URW (Unternehmensberatung Rubow Weber), was the dominant type design and production tool well into the 1990s.


Bell Centennial

Bell Centennial was designed by Matthew Carter to replace Bell Gothic as the typeface used by AT&T in its telephone directories. Bell Gothic had been designed by Chauncey H. Griffith for use on the Mergenthaler Linotype composing machine. In contrast, Bell Centennial was intended for use with the new generation of CRT phototypesetters. Although it was a digital design, it still involved lots of manual labor as Carter drew and inked-in each bitmap on quadrille paper before the letters were run through proofed with a Versatec plotter.



Chicago was one of a series of city-named bitmapped screen fonts designed by Susan Kare for the first Apple Macintosh. It was the most important since it was used for the operating system. Chicago was an original design while the other city fonts were “reasonable facsimiles” of familiar commercial typefaces: New York was derived from Times New Roman; Geneva from Helvetica; and Monaco from Courier. A smoothed TrueType version of Chicago was created by Charles Bigelow and Kris Holmes in 1990. Five years later, Charcoal, designed by David Berlow of The Font Bureau, replaced it as the operating system font for System 8.0. Yet, the original bitmapped Chicago remains one of the quintessential identifiers of Apple computers.

AMS Euler

AMS Euler was designed by Hermann Zapf with the help of Donald Knuth for the American Mathematical Society. It is intended for setting mathematics texts, which means that its basic roman character set is complemented by Greek, script and fraktur characters. AMS Euler was designed using METAFONT, a font manipulation program developed by Knuth, a Stanford computer scientist, in 1977.


Adobe Times Roman and Adobe Helvetica

Adobe PostScript fonts made their debut as resident fonts in the first Apple LaserWriter plain paper printer. They were licensed from Mergenthaler Linotype and the International Typeface Corporation (ITC), but digitized by Adobe. Although the digitization was poorly done—rumor has it that it was the work of friends and family of John Warnock and Charles Geschke, the founders of Adobe Systems—these fonts represented a significant jump in quality over existing laser fonts. Previously, resident fonts on laser printers were stored as bitmaps and thus could only be output at fixed sizes. In contrast, PostScript fonts were stored in an outline format that was filled in with bitmaps upon printing. They required less memory and were scaleable (capable of generating characters at sizes ranging from 1 point to 1000 points). The first LaserWriter contained only thirteen fonts: Times Roman (4 fonts), Helvetica (4 fonts), Courier (4 fonts) and Symbol.

The Adobe PostScript fonts were also sold independently for use on high-resolution devices. The same agreement between Adobe and Allied Linotype (the successor to Mergenthaler Linotype) that allowed Adobe to digitize Linotype faces permitted Linotype to install PostScript interpreters in its Linotronic 100 and 300 imagesetters. It was on these machines—with output resolutions of 1270, 2540 (and even 3300) dpi—that the defects of Adobe’s early digitization efforts became noticeable. On the LaserWriter, they were obscured by its coarse resolution.

The confluence of the Adobe PostScript language with the Apple Macintosh (introduced in 1984), the Apple LaserWriter and Aldus PageMaker—the first page layout program for the personal computer—led Paul Brainerd of Aldus to coin the phrase “Desktop Publishing” (DTP) to describe their joint potential.

Lucida Serif and Lucida Sans

Lucida, created by Charles Bigelow and Kris Holmes for Imagen, was the first original type family intended for laser printing. It was specifically designed with short sturdy serifs and low-stroke contrast to reproduce well when printed at low resolution (300 dpi). (Pellucida, a bitmapped companion intended for 72 dpi screen use, was released in 1986.) Lucida was also the first type family to revive Jan van Krimpen’s dream of a unified serif and sans serif design. This once-radical idea has since become fairly commonplace.

Lucida Sans Italic was also the first sans serif to have a true italic rather than an oblique, a practice that has been widely imitated since. (It was not the first chancery sans serif however. That honor belongs to Gerard Unger’s Flora [1980], originally designed to work with his Praxis [1977], both of which were issued by Hell. But Flora also functioned as a stand-alone face and that is how it has been viewed since ITC bought its rights and used it to inaugurate their Typographica series in 1989.)

Oakland, Emigre, Emperor and Universal

Oakland et al. were the first Emigre fonts. They were fixed-size bitmapped fonts. For example, Oakland comprised four fonts of increasingly refined resolution: Oakland 6, Oakland 8, Oakland 10 and Oakland 15. Zuzana Licko designed these four font families using FontEditor, a bitmap editing program. They were intended for output on an Apple ImageWriter, a pre-PostScript printer.

These first Emigre fonts were nothing special technologically or aesthetically. As bitmapped fonts they were inferior to Adobe’s PostScript fonts; and they were far simpler than existing bitmapped fonts such as Apple’s city fonts or the Fluent Fonts issued by Casady & Greene. What made them significant was their use in Emigre 3, a magazine that epitomized the promise of DTP even though it did not incorporate all of Brainerd’s elements. It was designed on a Macintosh using MacWrite and MacPaint; set in Licko’s fonts; and printed on an Apple ImageWriter. The issue also represented Emigre’s emergence as a type foundry as well as a magazine.


Licko redesigned her earlier Lo-Res Twelve bitmap font with straight line segments to approximate the “smooth printing” features of the Apple LaserWriter which processed 72 dpi bitmaps into 300 dpi bitmaps. The result was Citizen, Emigre’s first PostScript font.



After Citizen, the next step for Licko toward the design of a more traditional typeface was Matrix. It was designed with wedge serifs to keep the font data compact since early laser printers had limited (less than 1 MB) memory. The first Emigre fonts were optimized for low-resolution output rather than high-resolution output because they were for intended the magazine first and for other designers second. Despite this, Matrix became Emigre’s first widely adopted font.

TF Forever

The first original PostScript Type 3 font made using Fontographer was probably TF Forever by Joe Treacy. (The Altsys fonts were copies of existing typefaces.) Treacy began his design in 1984 using pencil and ink on paper before adopting Fontographer 1.0. TF Forever was the font that launched Treacyfaces, Inc., one of the first commercially independent digital type foundries.

Altsys Fontographer 1.0, the first type design software for the Macintosh, was released in 1985. With it one could make PostScript outline fonts, albeit of the Type 3 variety. Adobe dubbed its own PostScript fonts, and those made by its licensees, Type 1 and those made by others as Type 3. (Originally Adobe called them “user” fonts since they were not resident on PostScript printers. The earliest user fonts were created from specifications in the PostScript “Red Book” published in 1985) The latter lacked Adobe’s proprietary hinting technology and thus would not rasterize optimally when run on a PostScript printer. Nevertheless, Type 3 fonts represented the first assault on Adobe’s near-monopoly of digital type.

In the beginning, Altsys issued some fonts but they soon abandoned the idea once they discovered, as Adobe had, that properly digitizing letters was not as easy as it seemed. Instead, it was independent designers outside of the established type industry who succeeded in taking advantage of Fontographer’s capabilities.


ITC Stone Family

In the midst of their crash digitization of Linotype and ITC fonts in 1984, Adobe realized that the process was more difficult than expected. In order to improve Adobe fonts, Sumner Stone, then working at Camex, was brought in as Director of Typography. Although he had little impact on the quality of the first Adobe PostScript fonts, Stone turned Adobe from a software company that happened to make fonts into a true type foundry. One of his conditions of employment was that Adobe start a program of original type designs. The first Adobe original type design was the Stone family, licensed to ITC in 1988.

ITC Stone took the super-family concept one step further than Lucida. ITC Stone Serif and ITC Stone Sans were augmented by ITC Stone Informal. All three were designed to render well at both low- and high-resolution. ITC Informal—with its softened corners and cursive-derived “a” and “g”—was intended specifically as a substitute for typewriter fonts in the new office environment of personal computers and laser printers.

ITC Charter

Bitstream, established in 1981 by Mike Parker, Matthew Carter, Cherie Cone and Rob Friedman, was the first digital type foundry. Its fonts were initially created for use on dedicated Camex and Scitex workstations manufactured primarily for the newspaper industry. Its first three original fonts were Charter by Matthew Carter, Amerigo by Gerard Unger and Carmina by Gudrun Zapf-von Hesse. Carter began working on Charter in 1985. Like Matrix, it was designed to economize on font data—evident in its wedge-like serifs and abrupt branching—to accommodate limited printer memory. But by the time it was released printer memory had increased and its raison d’être no longer relevant.

Charter—which was licensed in 1993 by ITC as part of a GX joint venture with Bitstream—exemplifies the danger in designing a typeface for a specific technology or technological problem, something which Carter himself has warned against. One reason that Charter has managed to survive is that it has good bloodlines. Carter built it upon the armature of Pierre-Simon Fournier le jeune’s roman and italic.


In the late 1970s, Gerrit Noordzij began making typefaces for use in his book cover designs for the Dutch publisher Van Oorschots. He used a photolettering system of his own devising. In 1987 he used a prototype version of Ikarus-M to make Remer, a PostScript Type 1 font intended to replace his photo-titling designs. Since then Noordzij, the man whose teachings and students are behind much of the Dutch digital type revolution of the past twenty years, has designed a series of digital fonts aimed both at testing out his theories of letter design and for use in his own design projects. Ruse, released in 2000 though designed years before, is the only one of his fonts commercially available.


Compugraphic CG Type Library

When Compugraphic digitized its existing phototype library they designated the new fonts with the CG prefix. The fonts were not new, but the fact that they were created with Fontographer rather than Ikarus was significant. Compugraphic was the first established type foundry to adopt Fontographer.


Adobe dominated digital type between 1985 and 1989. During those years, the PostScript language was the de facto standard for printers interpreters. And Type 1 fonts were the only fonts that could be reliably output on PostScript devices. The licenses for both were carefully controlled by Adobe. Just as there were Type 3 fonts, there were clones, printer languages that emulated PostScript. The clones—which competed among themselves as well as with Adobe—were offered both by type companies and by printer manufacturers who were unhappy with Adobe’s licensing policies and fees.

Irritation with Adobe had been building for several years. It came to a boil in October 1988 when the Seybold Report on desktop publishing urged the clone makers to unite against Adobe. Seybold supported Fontware because Bitstream, its creator, had the best and most widely used type library. Fontware, announced in late 1987, was the first clone to “crack” the Adobe hinting code. But it was not until 1989 that the challenges to Adobe finally coalesced.

In May 1989, Microsoft acquired Bauer Enterprises and its PostScript clone. At the same time Apple announced its Royal font technology, an alternative to PostScript Type 1 fonts that had been in the works for two years. Adobe responded by announcing Adobe Type Manager (ATM), a program designed to generate on-the-fly Macintosh screen fonts using Type 1 fonts. In September, Microsoft announced that it had agreed to adopt the Apple Royal font technology and font library for use on its newly acquired clone. The Microsoft-Apple alliance threatened both Adobe PostScript and Adobe fonts. Under duress, Adobe announced that it would publish the specifications for its Type 1 font format as soon as possible. This was the end of encrypted Type 1 fonts.

Adobe Originals

Sumner Stone’s goal of a series of original fonts from Adobe was formalized in 1987 with the establishment of the Adobe Type Advisory Board—consisting of Roger Black, Max Caflisch, Alvin Eisenman, Stephen Harvard, Lance Hidy and Jack Stauffacher—and the beginning of the Adobe Originals program. The first fruits of the new program appeared in 1989 with the release of Lithos, Trajan and Charlemagne, a trio of display fonts designed by Carol Twombly, and Utopia and Adobe Garamond, a pair of text fonts designed by Robert Slimbach. Utopia, the only one of these fonts not to be historically-based, was the most original of the first group of Adobe Originals.

Many graphic designers in the 1980s, having begun their careers in the era of metal type and having never fully embraced phototype, were very skeptical of digital type and its promises. They complained about the quality (and provenance) of typefaces, the lack of large character sets needed to properly do book typography, and the acceptance of one-sizes-fits-all master designs—all practices inherited from phototype. (They also complained about the “jaggies”—the coarseness that many digital fonts displayed at large sizes—but with ATM for screen fonts and PostScript for printer fonts jaggies were no longer a signficant problem.)

Adobe Garamond was the font that changed their minds. It was a real Garamond, based directly on the types of Claude Garamond, rather than a false one like ITC Garamond and the other so-called Garamond revivals that actually derived from the types of Jean Jannon. At the insistence of the Adobe Type Advisory Board, whose members had a strong interest in book design and text typography, Adobe Garamond (and Utopia) included an expert set (consisting of old-style figures, small capitals, ligatures, piece fractions and alternate characters). As a partial solution to the lack of optical scaling in digital type, Adobe Garamond included both regular (text) and display versions of its basic roman and italic.

The careful historical research behind Trajan and Adobe Garamond and the addition of typographic refinements to Utopia and Adobe Garamond indicated how serious Adobe was about issuing quality typefaces (a tacit acknowledgement that their early fonts were flawed). The Adobe Originals program marked the coming of age of Adobe as a type company and the broad acceptance of digital type by graphic designers. (They may also have contributed to the support Adobe gained in its subsequent fight with Microsoft and Apple.)


The first typeface from Letterror, the Dutch duo of Erik van Blokland and Just van Rossum was a RandomFont they dubbed Beowolf. It was digitized by hand in Ikarus M (a version of Ikarus developed for the Macintosh by Petr van Blokland) and then programmed to change at random during printing. Beowolf could metamorphose from Dr. Jekyll, a mild-mannered font with a pleasing roughness suggestive of letterpress printing, to Mr. Hyde, a wild-eyed and jagged font. It challenged the reigning notion that outline fonts were inherently superior to bitmapped ones. When Beowolf was subsequently released by FontShop International, it was tamed. In place of a programmed RandomFont there was now Beowolf 21, Beowolf 22 and Beowolf 23, a trio of fixed fonts of increasing distortion.

Rotis Family

Rotis was designed in 1988 by Otl Aicher as an attempt to forge a new type family that combined the best attributes of Times New Roman and Univers: legibility, functionality and economy. The result was four variants of roman and grotesque—Rotis Sans, Rotis Semi-Sans, Rotis Semi-Serif and Rotis Serif (originally designated R1, R2, R3 and R4 respectively)—that share the same set width. Rotis Serif is an old-style roman; Rotis Semi-Serif is Rotis Serif without the bracketed serifs at the bottom of letters; Rotis Semi-Sans is a grotesque, but with thick-and-thin strokes in the manner of Optima; and Rotis Sans is a traditional grotesque. The Rotis family, released by Agfa Compugraphic (the successor to Compugraphic) in 1989, was a more logical extension of the Lucida serif/sans model than ITC Stone.

WTC Our Bodoni

Massimo Vignelli has argued for several decades that graphic designers only need a handful of typefaces. His preferences have been Bodoni, Century Expanded, Garamond no. 3, Times Roman and Helvetica. In 1989, he decided to redesign each of these typefaces to have the same set width and x-height of Helvetica so that they could be swapped in and out of a layout without effecting the line breaks or the text flow. The only one of these typefaces to ever appear was WTC Our Bodoni designed in conjunction with Tom Carnase of the World Typeface Corporation. The ascendancy of page layout programs like PageMaker and QuarkXpress, with their ability to effortlessly flow and reflow text, made Vignelli’s idea irrelevant and none of the other proposed redesigns were carried out. WTC Our Bodoni since been largely forgotten amidst the welter of other Bodonis.

1989, 1990

Arcadia (formerly called Campanile), Industria and Insignia (formerly called Stadia)

The first so-called “rock star” graphic designer was Neville Brody. He first gained fame in the 1980s with his work for Fetish Records and for The Face and Arena magazines. His 1988 one-man show at the Victoria & Albert Museum in London—a rare occurrence for a living designer—cemented his status. In the wake of that show, Linotype Library (formerly Allied Linotype) attempted to capitalize on Brody’s fame with the release of digital fonts based on custom typefaces—originally made in the pre-digital manner using graph paper, compass and ruler, and photostat cameras—created for The Face and Arena. Although type foundries have been swiping ideas from artists and designers for a century (viz. Otto Eckmann, George Auriol and Cassandre), the release of Arcadia, Industria and Insignia marked the moment when the New Wave typography of the 1980s made it into the typographic mainstream. Not surprisingly, by the time the mainstream had caught on to Brody, a new design “rock star” was on the horizon.


FF Scala

Martin Majoor designed Scala in 1988 for the Vredenburg Centre in Holland. That same year, FontShop, Berlin was founded by Erik and Joan Spiekermann. Neville Brody was added as a partner in 1989 as FontShop, Berlin grew into FontShop International (FSI), a global network of font distributors. FSI provided an outlet for the new breed of independent digital type foundries that had begun to appear. FF Scala was the first FF (FontFont) text face released by FSI. It was the first of the new wave of Dutch types that have reinvigorated type design in the past fifteen years.

Keedy Sans, Template Gothic and Dead History

Following Emigre’s release of Zuzana Licko’s designs, graphic designers and design students began submitting fonts to the magazine. In 1990 Emigre released several of these designs, marking its transition from magazine to foundry. In various ways they all challenged the assumptions underlying traditional type design.

With its mix of rounded and sheared stroke endings, Keedy Sans, designed by Jeffery Keedy of Cal Arts (California Institute of the Arts), questioned the notion of consistency as a feature of a good typeface. Another typeface from Cal Arts was Barry Deck’s Template Gothic, inspired by a degraded stenciled sign in a laundromat. It represented the newfound interest in the vernacular that had arisen alongside—and often in opposition to—the computer in the 1980s. Template Gothic was Emigre’s first widespread hit. Dead History by P. Scott Makela of the Cranbrook Academy of Art (with help from Zuzana Licko)—a blend of Linotype Centennial and VAG Rundschrift (VAG Rounded) exemplified the new “sampling” approach to type design that Fontographer had made possible. Its name personified the attitude of the younger designers toward traditional type design.


Following Citizen and Matrix, the next stage in the development of Zuzana Licko as a type designer was Journal, a font designed to recapture the look of metal types printed letterpress when printed at low resolution. To accomplish this look, Licko formed curves from straight lines (vectors), much as she had with Citizen. However, Journal—unlike Citizen—looks more like a traditional serif typeface.

Type Before Gutenberg 1: Herculanum, Omnia and Duc de Berry

The “Type Before Gutenberg” series of fonts from Linotype Library is brilliantly named but typographically odd. They consist of digital “revivals” of letterforms from Antiquity and the Middle Ages that were never typographic in the first place but, instead, were either inscriptional or calligraphic. The first of these sets included Herculanum by Adrian Frutiger (based on 1st century Old Roman Cursive), Omnia by Karlgeorg Hoefer (a Roman uncial), and Duc de Berry by Gottfried Pott (a 15th century northern French bâtarde). The anachronism of Type Before Gutenberg 1 exposed the leveling influence of the digital world in which the distinctions among handwriting, calligraphy, lettering, typography and other modes of letter making have dissolved.


Hard Times and Fudoni

Fontographer 3.1 (released in 1991) was the first version of the software program with the ability to create Type 1 fonts. This was made possible by Adobe’s 1989 decision to end the encryption of its PostScript Type 1. This meant that type designers could not only use Fontographer to design fonts, but that they could also use it to dissect and alter existing fonts. Two fonts that exemplified this new-found ability were Hard Times and Fudoni.

Jeffery Keedy described his design of Hard Times as an “ironic commentary” on classic typefaces. Whether the font was truly ironic—its name conjures up not only a typeface but also Charles Dickens and the Great Depression—or just the result of someone goofing around with Fontographer is immaterial. Either way, Keedy’s alteration of Times New Roman—hacking off and reassembling serifs and other parts—encouraged a whole slew of typographic mutilations in the 1990s (including fonts “designed” for advertising campaigns for Putnam Investments and Air France).

The collage approach to type design underlying Dead History was made explicit in Max Kisman’s Fudoni, a Frankenstein-like grafting of Futura and Bodoni together that made no attempt to hide its sutures. Whereas Hard Times took a subtractive approach to altering existing typefaces, Fudoni took an additive one.

Not Caslon

Hot on the heels of Adobe Garamond came Adobe Caslon (1990). Adobe had successfully kicked off an interest in reviving the classic typefaces of the past that continues today. Not Caslon, designed by Mark Andresen and issued by Emigre, was a satiric riposte to the idea of type revivals, a concept that has always been controversial. Not Caslon is an amalgamation of body parts from Caslon 540 cut up and put back together differently. Depending upon your sensibilities, the result is either a Guernica-like display of typographic horrors or a blackly humorous boneyard for a dead typeface.

FF Erikrighthand and FF Justlefthand

These two complementary typefaces were based on the cancellaresca-influenced handwriting of Erik van Blokland and Just van Rossum, the Letterror “twins.” The equivalents of Mistral in the metal era, they were the first digital handwriting fonts to become wildly popular.

The release of FF Erikrighthand and FF Justlefthand coincided with an interest amongst the general public in fonts based on one’s own handwriting. This interest was ignited by the introduction of Signature Software Technology in 1988 and was fanned by a rash of individuals offering personal handwriting fonts (complete with randomizer program) “for just $99.” The first of these handwriting font companies was The Chank Company, established by Chank Diesel in 1992 and written up in The Wall Street Journal in 1997. Font design had truly become democratic.

FF Blur

Adobe introduced Photoshop 1.0 in 1990. Although designed for the manipulation and reproduction of photographs, it was possible to use Photoshop to set and edit type. Fonts were once again pixelated images. Neville Brody took advantage of this to run Helvetica through Photoshop’s blur option and then vectorized the results. Voilà! FF Blur, a program-based font whose biomorphic features challenged the reigning notion that such type should look bitmapped or constructivist.

A*I Prospera II

Peter Fraterdeus, founder of Alphabets, Inc., one of the earliest independent digital type foundries, began Prospera in 1985. A*I Prospera was released in 1987 as a Type 3 font and revised—following the release of Adobe’s encryption codes—as a Type 1 font called A*I Prospera II in 1991. It was the first original digital typeface to be re-engineered following the logic of type as simply a specialized form of software subject to the same periodic upgrades as programs such as Adobe Illustrator or Microsoft Word.



Poetica can be considered to be the first italic type family. Designer Robert Slimbach has revived that brief moment at the opening of the 16th century when italic was an independent type style rather than an adjunct to roman. Poetica I consists of Chancery I, II, III and IV; Chancery Expert, Roman Titling, and Roman Titling Alternates. Poetica II includes Swash Capitals I, II, III, and IV; Initial Swash Capitals; Lowercase Alternates, Beginnings and Endings; Ligatures, Ampersands and Ornaments. This overabundance of options—many of which are unnecessary—obscures the true significance of Poetica. For the first time since Arrighi, hierarchical texts can be set in italic.


Originally designed for use in a Dutch dictionary, Lexicon by Bram de Does was the first comprehensive text type family. It consists of two main branches: Lexicon No. 1 (short ascenders and descenders) and Lexicon No. 2 (normal ascenders and descenders) each with twelve variants (six roman and six italic weights) that themselves have five sub-variants (old style figures, tabular old style figures, tabular lining figures, expert set—small capitals and tabular old style figures, and a pi font with superior and inferior figures and special diacritics). The concept of variable ascender and descender length had previously been explored by de Does in Trinité (1982), a photocomposition face.

Lushus and Caustic Biomorph

The culmination of the experimental typographic ferment of the last half of the 1980s was the launching of Fuse in 1991. The brainchild of Neville Brody and Jon Wozencroft, Fuse was a “fontzine,” a series of four experimental fonts designed around a given theme, accompanied by four A2 posters showing them in action. Fuse was seen as an outlet to allow type designers to “challenge conventional thinking about the form and function of typography.”

The theme for Fuse 4 was Exuberance and two of the fonts were Lushus by Jeffery Keedy and Caustic Biomorph (has any typeface had a more memorable name?)by Barry Deck. These two typefaces seemed to embody all of the aesthetic horrors that established designers found objectionable about the new wave of “experimental” digital fonts. Lushus, with its Tuscan bifurcated serifs and other Victorian decorative excesses, was guaranteed to offend both traditionalists and modernists who considered the 19th century to be the previous typographic nadir. Whereas Lushus was sarcastically “pretty,” Caustic Biomorph was downright ugly, the apparent result of an industrial/chemical accident. It provided proof for those who thought that typography had reached an apocalypse.


The font wars between Adobe and the Apple/Microsoft alliance heated up with the release of Windows 3.1 in 1992. It was the first program to include TrueType fonts rather than PostScript ones. Its thirteen core fonts—all developed for Microsoft by Monotype—were, in alphabetical order: Arial, Bookman Oldstyle, Book Antiqua (Palatino in disguise), Corsiva, Century Schoolbook, Century Gothic, and Times New Roman. Arial, designed in 1988, has since become one of the most widely-used fonts thanks to its resemblance to Helvetica—Monotype deliberately designed it to match the set widths of PostScript Helvetica—and to its presence in Microsoft products.


The 1980s postmodern backlash against modernism inevitably led to a rebellion against Helvetica, the typeface that had personified corporate modernist typography for two decades. One solution to Helvetica’s worldwide ubiquity was a typeface as far removed from “objectivity” as possible: Remedy, designed by Frank Heine and released by Emigre. With its curlicues, jaunty baseline, random ornamentation, and general ungainliness, Remedy appeared to be nothing more than a doodle. It was personal without being handwriting.

HTF Didot

During his tenure as art director of Harper’s Bazaar from 1934 to 1958, Alexey Brodovitch made Bauer Bodoni synonymous with elegance, grace and sophistication. In an attempt to recapture those glory years, art director Fabian Baron commissioned a new Didot type from the young Jonathan Hoefler in 1991. Hoefler’s design was christened HTF Didot and was released to the public a year later, coincidentally with two other Didot revivals: Linotype Didot by Adrian Frutiger and LP Didot by Garrett Boge. What set HTF Didot apart from its rivals was its range of optically-adjusted master sizes—six, eleven, sixteen, 24, 42, 64, and 96 point respectively—that preserved the fragile hairlines so typical of the Didot style. The 96 point Didot Light was the best proof—if any was needed by this time—that the jaggies were no longer a concern with digital type.

Myriad MM and Minion MM

The significance of the optical scaling in HTF Didot was overshadowed in 1992 by the release of Myriad MM and Minion MM, the first Multiple Master fonts from Adobe. Multiple Master typefaces applied the concept of interpolation—a key feature of Ikarus, METAFONT and other digital type design tools—to Type 1 fonts enabling users, via the PostScript language, to generate on-demand versions of a given font intermediate between its prepared master design “instances,” The master designs of an MM typeface determined its dynamic range along one or more axes. The axes could be weight, width, optical size or style.

Myriad MM was originally developed by Robert Slimbach and Carol Twombly as a generic typeface for use in Adobe Acrobat. (Version 1.0 was released in 1993.) It is a two-axis (weight and width) sans serif that redefined the concept of a type family. The three-axis Minion MM, an updated version of Slimbach’s earlier Minion (1990), was the first Multiple Master typeface to have optical scaling capabilities. With its expert sets, it was the first digital font to achieve the goal, set by the Adobe Type Advisory Board, of matching the capabilities of metal type.



Sumner Stone developed Cycles, based on his earlier Stone Print (1991), for use specifically at 9 point size in Porter Garnett: Philosophical Writings on the Ideal Book designed by Jack Stauffacher. Whereas Stone Print was designed as a magazine font, Cycles was intended as book face. Consequently, Stone has since expanded it to include additional size-dependent variants: 11, 18, and 48 point sizes in 1997; and 7, 24 and 36 point sizes in 2004. Stone’s ongoing project—which also includes a display version called Arepo (1995)—hearkens back to the days of the punch cutter, who refined a single design over and over again.

FF Kosmik

One of the advantages of digital type over previous technologies has been the ability to include extra characters—alternates, ligatures and swashes—in a font at minimal or no extra cost. But easily accessing and using such characters has been difficult. Erik van Blokland’s FF Kosmik, with three alternates for each character, solved the problem by (randomly) automating the substitution process via his Flipper software. As a result, van Blokland dubbed FF Kosmik a FlipperFont.

The Proteus Project: Leviathan, Ziggurat, Acropolis and Saracen

Massimo Vignelli’s vision of a group of fonts with identical set widths finally came to fruition with The Proteus Project by Jonathan Hoefler. Ironically the four typefaces that comprised The Proteus Project—Leviathan, Ziggurat, Acropolis and Saracen—are derived from wood type styles that modernists have always viewed with horror. Leviathan is a grotesque (sans serif), Ziggurat is an Egyptian (slab serif), Acropolis is a Grecian (chamfered serif) and Saracen is a Latin (wedge serif). The logic of this quadruplexed family is perfectly in keeping with the pantograph-based mechanical production of most wood type.

Architype Series 1: Architype Bayer, Architype Bill, Architype Van Doesburg, Architype Van der Leck and Architype Renner

From William Morris to Stanley Morison to Adobe, typographic revivals have always been part of the traditionalist strain in typography. The emphasis has consistently been on the “classic” metal typefaces of the pre-pantograph era. But, in the wake of renewed interest in the “pioneers of modern typography” (as Herbert Spencer labeled them in his 1969 book of that name), it was not surprising that someone finally created modern typographic revivals. The true genius of the Architype Series from David Quay and Freda Sack of The Foundry was that they revived alphabetic experiments rather than actual typefaces. The first of these sets—two others followed in 1995 and 1999—included Herbert Bayer’s Universal Alphabet and Paul Renner’s radical design for what eventually emerged as the beautiful, but tame, Futura.


Hoefler Text

From a typographic standpoint the most exciting aspect of System 7.5 was the inclusion of QuickDraw GX Type, software that promised the ability to intelligently substitute characters—small caps, ligatures, swashes and other special sorts (including mathematical symbols and foreign language characters)—in text, distinguishing between semantic and stylistic contexts. TrueType GX fonts also had the potential to contain font variations similar to those in Adobe’s Multiple Master fonts. Unfortunately, GX flopped because it lacked multi-vendor support and a critical mass of GX types that could display its capabilities. One of the few that did do this successfully was Hoefler Text, one of three original fonts—the other two were Apple Chancery and Skia—included with System 7.5. Hoefler Text, designed by Jonathan Hoefler, was an oldstyle typeface that included various weights as well as small caps, old style figures, ligatures, swash characters and ornaments.

Adobe Jenson

Adobe Jenson by Robert Slimbach was the first type revival with Multiple Master features, notably optical scaling. It was based—like so many types in the Arts and Crafts era—on the quattrocento roman of Nicholas Jenson.

HTF Historical Allsorts

The question of what constitutes an authentic historical revival has been a source of controversy since at least the early 1920s when Monotype twice revived the types of Francesco Griffo, first as Poliphilus and then as Bembo. Poliphilus was a faithful redrawing of the roman in the Hypnerotomachia Poliphili (1499), while Bembo was a recreation of the spirit behind the roman in De Aetna (1495). Ever since, historical revivals—including the Adobe Originals—have tended to follow the lead of Bembo rather than Poliphilus. HTF Historical Allsorts returned to the Poliphilus model. The four fonts that constitute the set—Fell Type Roman, English Textura, Great Primer Uncials, and St. Augustin Civilité—were autotraced by Jonathan Hoefler from printed specimens and only minimally retouched afterwards.

Penumbra MM

Lance Hidy’s Penumbra was the first Multiple Master font to include a style axis alongside one for weight. Its style axis focused on the amount of serif present. The four basic instances ranged from sans serif to fine serif to semi-serif to serif. Thus, in one design—albeit a titling face—Penumbra MM achieved what the Rotis family had set out to do.

Thesis Superfamily

In the wake of the Lucida, Stone and Rotis families it was inevitable that a designer would follow the logic of the superfamily—a type family that includes stylistic variants as well as the traditional ones based on slope, weight and width—to its ultimate conclusion. That designer was Luc(as) de Groot and the superfamily is Thesis.

The first three branches of the Thesis family were FF TheSerif, FF TheSans and FF TheMix (sans serif capitals with a semi-serif lowercase). Each style came in eight weights (extra light, light, semi-light, normal, semi-bold, bold, extra bold and black) and each of those weights had six variants (regular, small caps, expert set, italic, italic small caps, and italic expert set) making for a total of 144 fonts. As if this were not enough de Groot has continued to add to the Thesis family over the past decade: TheSans Monospaced (a fixed-pitch alternative to Courier), TheSans Monospaced Condensed, TheSans Typewriter (proportionally spaced), TheAntiqua (a more traditional serif face than TheSerif) and so on.

(The Lucida “family” of fonts runs a close second to Thesis in size, but it is fundamentally different. Whereas the different branches of Thesis share a common skeletal form, that is only true of the original Lucida fonts and some of the later ones [e.g., Lucida Bright or Lucida Typewriter]. Others, such as Lucida Handwriting and Lucida Blackletter, have nothing in common beyond their brand name. At best, they are the cousins, in-laws and stepchildren of Lucida and Lucida Sans.)



Jeremy Dean created Crackhouse by rubbing down a Letraset sans serif typeface and then repeatedly lifting up the sheet several times so that the letters “cracked.” The distressed font was an instant hit, typifying the “grunge” typography of the time. It became the cornerstone of Bad Neighborhood, the first theme-based font collection from House Industries.

Typefaces like Crackhouse on one hand and HTF Didot on the other—ranging from the ridiculous to the sublime—made it abundantly clear by the mid-1990s that digital type could take on any appearance, that there was no such thing as a digital “look.”


The Walker Art Center in Minneapolis commissioned Matthew Carter to design an institutional typeface for it. Carter responded with an ordinary, all caps sans serif that came with five varieties of what he termed “snap-on” serifs, as well as overscores, underscores and ligatures. The different serifs—hairline, bracketed, wedge and two kinds of slab (half and full)—provided Walker with the ability to continually change its look. It is a single typeface with multiple personalities, embodying Carter’s long-standing claim that digital type differs from the type of the past in its mutability.

Just as early metal typefaces were influenced by the broad-edged pen, so too have digital typefaces been influenced by a tool: Fontographer with its cut-and-paste abilities. Cloning of letter parts has become a standard working practice among digital type designers. To some extent Walker’s snap-on serifs have turned the graphic designers of the Walker Art Center into type designers.

Base-9 and Base-12

With the growing popularity of the World Wide Web and the internet by the mid-1990s, type designers were again confronted by the issue of resolution, this time on screen. In 1995, Zuzana Licko tackled the subject with the design of Base-9 and Base-12, a family of screen fonts (both sans serif and serif) with companion printer fonts. She reversed the usual process of adapting printer fonts to the limitations of coarser screen fonts and designed the screen fonts first. They were designed to function at their best when set at 9 or 12 point respectively (or at multiples of each). In addition, they have consistent spacing both on screen and when printed.


Verdana and Georgia

Matthew Carter designed both the sans serif Verdana and the serifed Georgia as screen fonts in 1996. Unlike Licko’s Base-9 and Base-12 they are not bitmap-based. (The ancestor of Verdana is Bell Centennial.) Microsoft offers them free to anyone as downloadable fonts for use on websites and personal computers, though they can be used with success in printed documents as well. They are intentionally ordinary to further their widespread acceptance and help guarantee a long shelf life. Consequently, they may become the new Helvetica and Times Roman.

Mrs. Eaves

With the design of Mrs. Eaves, Zuzana Licko finally joined the ranks of the traditional type designers. A fanciful interpretation of Baskerville rather than an attempt at a faithful revival, the font still has some tricks up its sleeve. Licko has included “petit caps” as well as the usual small caps and caps; and, with the programming help of Letterror, she has augmented the font with a wide range of “Smart Ligatures” that go beyond the familiar fi, fl, ffi and ffl. Yet, the normalcy of the basic Mrs. Eaves design seemed to mark the end of the “experimental” period in digital type design that Emigre had kicked-off a decade earlier.

HTF Fetish no. 338

A sure sign that the “experimental” period in digital type design was indeed over was the release in 1996 of Jonathan Hoefler’s satiric HTF Fetish no. 338. The tongue-in-cheek named font was designed as a response to the popularity of mongrel fonts such as Exocet. HTF Fetish no. 338 allegedly appropriated Gothic, Victorian, Byzantine, Celtic and Moorish forms.



Among the fonts included in Fuse 15—Cities was Bits by Paul Elliman. Composed of objects found in the street it was the first true vernacular font. (Elliman has since continued to “upgrade” the font as he finds new “bits” to replace existing ones.)


ITC Founders Caslon

The issue of the validity of autotraced type revivals raised by HTF Historical Allsorts was reignited with the release of ITC Founders Caslon in 1998. Justin Howes, the font’s “designer,” letterpress-printed samples of each size of Caslon type and then scanned the results. He left in the irregular edges, making no attempt to smooth or regularize the designs. The ITC release included 12, 30, 42 and 96 (poster) master point sizes. Later, under the venerable name H.W. Caslon (the rights to which he had purchased), Howes re-issued these fonts along with eleven others—Caslon 1776; Founders Caslon Display (8, 10, 12, 14, 18 points); and Caslon Display (22, 24, 30, 36, 42, 48, 60, 72, and 96 points)—representing the complete range of Caslon type that survived into the 20th century.

The results of Howes’ efforts are curious. The fonts are undoubtedly authentic, but when printed offset on modern paper they often look quaint. Caslon’s spirit is elusive.


Syntax LT

Syntax was designed by Hans Ed. Meier as a sans serif equivalent of a Garamond. He began work on the design in the mid-1950s, and it went through several iterations before Stempel finally released it in 1968. (It was the last metal typeface from the foundry.) Along the way, Meier was forced to make compromises in the shapes of certain letters to accommodate the needs of hot metal casting. Further compromises were made when Syntax was converted for photocomposition. These compromises were perpetuated when it was digitized in the 1980s. Long a cult favorite, Syntax gained widespread popularity in the 1990s, prompting Meier and Linotype Library to revise and expand the font. Syntax LT (as the new version is called) is the first (and probably only) metal typeface to be digitized by its originator. As such, it is arguably more authentic than its predecessor. Yet, for many who have worked with the original Syntax for years, the new version is not an unqualified improvement, prompting the question as to whether the type designer or the type user is the ultimate judge of a typeface’s utility.

Warnock Pro and Silentium Pro

In a bid to end the “font wars” of the early 1990s, Adobe and Microsoft collaborated on a new font format called OpenType. OpenType is an extension of TrueType with support for PostScript font data. The OpenType format addresses several goals: broad multi-platform support, better protection for font data, smaller file sizes, handling of large glyph sets using Unicode encoding (allowing for better support of international character sets), and more advanced typographic control. As such, OpenType goes beyond the unfulfilled promise of GX fonts.

Although the first specifications for the OpenType format were published in 1997, it was not until 2000 that the first OpenType fonts were released. They were bundled with InDesign 1.0, Adobe’s page layout program developed as a rival to QuarkXpress. Most of these fonts were simply updated versions of existing Adobe fonts such as Minion. The first original typefaces designed explicitly for the OpenType format were Warnock Pro by Robert Slimbach and Silentium Pro by Jovica Veljovic.


Lucida Grande

With the release of System 8 for the Macintosh David Berlow’s Charcoal replaced Lucida as the Macintosh’s operating system font. Lucida returned in the guise of Lucida Grande with the advent of System OSX in 2001.



Peter Bil’ak designed Fedra Serif A and Fedra Serif B in 2003 to accompany Fedra Sans (2001). Fedra Serif A has a tall x-height and shortened descenders while Fedra Serif B has longer ascenders and descenders along with stronger stroke contrast. With the pairing of Fedra Serif and Fedra Sans Bil’ak has combined the notion of a large multi-style type family (e.g., Thesis) with that of an adjustable text font (e.g., Lexicon).


Can a typeface communicate what’s special about a city? This was the question the Design Institute of the University of Minnesota put to six typography studios in 2002 as part of a broad concept to brand a festival in celebration of the Twin Cities of Minneapolis and St. Paul. The winner of the Design Institute’s competition was Twin by Letterror (Erik van Blokland and Just van Rossum), the first typeface “capable of responding, in real time, to dynamic urban conditions such as wind and temperature.” With accompanying online software (grandiosely—and humorously—called the Panchromatic Hybrid Style Selector), text set in the 300+ character Twin can change from “formal” to “round” to “weird” as the temperature changes and can billow along with the wind.



In 2004, Stone Type Foundry issued Magma, a humanist sans serif along the lines of Gill Sans. Sumner Stone has billed it as the first typeface with Halo fonts (as Stone has christened them). Essentially, Halo fonts are subtly heavier versions of regular size-specific fonts that will retain their optical weight when reversed out. They also can be used at very small sizes to insure readability.


Auto, a “triple-italic sans serif”, was designed by the Dutch/Finnish design studio and type foundry called Underware as a means of calling attention to what they saw as “an unexplored palette.” The font’s three italic options—described as “formal,” “flavourable” and “impress your grandpa with this one”—promise the user the option of setting quotes within quotes, identifying quotes within spoken text and distinguishing speakers in a play.

Although Auto is not a particularly auspicious design, it signals something new in digital type—a complex font in search of a problem to solve rather than the other way around. In the past complex fonts (such as Lexicon or Poynter) were designed in response to specific client briefs. Auto is not outwardly experimental like Dead History and its ilk from the early 1990s. Its experimentation comes from within the tradition of type design rather than from outside.


Corporations, newspapers, magazines and even museums have long commissioned custom typefaces as part of identity systems. Now universities have been added to that list. Last year Matthew Carter designed an updated Aldine for Yale University named, naturally enough, Yale. The Yale family includes variations geared for signage, print and web work. It is available free of charge to all Yale administration personnel, faculty and students.

What will happen next with digital type?

Here are some thoughts that the making of this list has sparked.

It has taken a while, but graphic designers have finally begun to understand and accept OpenType. Adobe has forced the issue by including OpenType fonts with InDesign and aggressively linking InDesign to its entrenched Photoshop, Illustrator and Acrobat programs. What will happen once InDesign fulfills Adobe’s goals and successfully kills off QuarkXpress? Will Adobe then have a monopoly on the tools used by graphic designers? (Adobe has just purchased Macromedia giving it control over Flash, Dreamweaver and Director.)

The more software programs that Adobe controls, the more it controls the venues in which type appears. And that will make it likelier that PostScript 1 fonts will soon disappear. Adobe has already embarked on a program to convert all of its PostScript 1 fonts to the OpenType format. The major old-line type companies like Linotype Library and Monotype are following suit as are leading independent type foundries such as the Dutch Type Library and The Foundry. But will individual type designers—especially those just dabbling—be motivated to do the same to their PostScript Type 1 fonts? If not, does that mean we are entering a new age of consolidation in which only professional type designers design type?

One factor that may hasten such a scenario is the loss of inexpensive, widely available, easy-to-use font design programs like Fontographer and FontStudio. FontLab, the preferred program among professionals today, is costlier, not widely distributed and requires more effort. It is a serious tool rather than a toy. (There is now a cheaper version called TypeTool, a light font editor that can handle simple font manipulation, but not fully build a font.) Furthermore, where once Letterror was unique in its embrace of programming as part of type design, other type designers are becoming more and more comfortable with scriptwriting. Then again, maybe Adobe will revive Fontographer now that it has acquired Macromedia. Or perhaps it will incorporate features from Fontographer into Illustrator and give that popular program the ability to create OpenType fonts.

If type design once again becomes the province of the professional, then we will see an end to the era of dirt-cheap prices. As it is, foundries like The Enschedé Font Foundry, the Dutch Type Library and Hoefler & Frere-Jones Typography charge considerably more for their fonts than do resellers like Phil’s Fonts or FontShop. (Though it should be remembered that the fonts from these companies are still incredibly cheap by the standards of the pre-Macintosh era of type shops.) A rise in the price of fonts is long overdue. Perhaps we will no longer buy fonts, but will instead return to the era of the type shop. House Industries is planning a Photo-Lettering Inc. website in which customers can type in their text, choose a font or fonts to set it in, pay a per-word fee, and in return receive a PDF version of their layout. No fonts will be purchased. For the graphic designer, such a transaction will be cheaper than buying a font, yet the cumulative impact of many of these transactions will be more remunerative to the type foundry. From the perspective of the type foundry, this proposed service would also reduce font piracy.

While House Industries’ website is still in the planning stages, the way in which we use fonts has already been seriously affected by Acrobat. PDF documents have become as crucial to Adobe in the world of the internet as PostScript has been in the world of print. By enabling anyone equipped with the free, downloadable Acrobat Reader to see a document exactly as it was created—whether or not they have the fonts used in its making or not—PDFs have significantly reduced the number of people who need to buy fonts. This is especially true now that the just-released Acrobat 7.0 allows recipients to edit PDFs. We may find ourselves back in a world where the only people who buy fonts are graphic designers. Already, the average computer user has little incentive to buy fonts since so many come bundled with software programs.

Although the word “experimental” bothers me, Heller may be right in his sense that an era of experimentation in type design is over. Increasingly, the new type designs that I find exciting are those that deal with the traditional concerns of legibility, readabililty, economy, flexibility and transparency (that is, typefaces that seek to solve typographic problems rather than serve as outlets for artistic innovation). Since the late 1990s, graphic designers have increasingly embraced such fonts—witness the popularity of Thesis and Miller—because they enable them to deal quickly, easily and reliably with complex information in a wide variety of media. Dutch type designers have been in the forefront of this trend, which is likely to accelerate as the number of OpenType fonts grows and InDesign gains on QuarkXpress.

It may seem as if type design is more conservative than before, but when one looks more closely at the digital fonts of the past, those with staying power have been essentially conservative—that is the inherent nature of type. Lucida, ITC Stone, Minion, Meta, Scala, Mrs. Eaves and Lexicon that have remained popular while Template Gothic, Exocet, Remedy and Lithos were nothing more than shooting stars. In the end, the “experimental” fonts of the digital era were no more experimental than many of the type designs that sprouted during other periods of technological change—whether the 19th century or the 1960s. If anyone has any doubts about this, I urge them to check out Centralschrift, Crystal, Mikado, Epitaph (revived in 1993), Confetti, Huit (a forerunner of Blur), LED-7, Dimple Right or Block-Up.

Researching this list proved to be more difficult than anticipated. I found that my memory of events was not reliable. So, since the subject was digital type it seemed natural to turn to the internet for information. Unfortunately, much of what I found there was wrong, misleading or contradictory. I did gain useful facts from a number of sites including,, Emigre, the Dutch Type Library, The Enschedé Font Foundry, Monotype Imaging and several dedicated to the history of Apple, TrueType and other geeky topics. But I often had to corroborate what I learned on the internet with print sources from the 1980s and 1990s, as well as with conversations with type designers and others who were there. I want to thank Sumner Stone, Matthew Carter, Joseph Treacy, Jonathan Hoefler, Erik Spiekermann, Garrett Boge, Gerrit Noordzij, James Montalbano and Frank Romano taking the time to answer my questions and provide feedback. But, as usual, all the opinions expressed here are entirely mine.

Tags Article Voice typography

Paul Shaw is a calligrapher and typographer working in New York City. In his 20 professional years as a lettering designer, he has created custom lettering and logos for many leading companies, including Avon, Lord & Taylor, Rolex, Clairol and Esté Lauder. Paul has taught calligraphy and typography at New York's Parsons School of Design for over 10 years and conducted workshops in New York and Italy. His work has been exhibited throughout the United States and Europe.