The font wars
Typography is what language looks like
Ellen Lupton, 2010
“The Font Wars” is a period of time between the 80’s and the 90’s in which different technology companies battled to dominate the font technology market. Fonts were fundamental to the success of desktop publishing. Desktop publishing was a revolution in the design and production of documents, it allowed regular people and small businesses to produce and print documents at a lower cost than what was possible with any other printing technology at the time.
Suzanne Crocker (2019) argues that scalable fonts, together with laser printers and software such as Pagemaker, afforded users to interact with an image on-screen which resembled how the finished product would look like, also known as what you see is what you get, a revolutionary concept which still influences the way in which humans interact with computers.
The Font Wars are often reduced to a business and technology dispute between technology giants Apple, Microsoft and Adobe to get TrueType or Adobe Type 1 fonts to be the standard for the industry (Herrmann, 2020). But the full story of this period reveals a diversity of actors, business decisions and technological developments which led to the democratisation of desktop publishing and literacy around the world. Charles Bigelow (2020) argues that more than a business dispute, the Font Wars were a manifestation and translation of ideas into software. These diverse ideas emerged from areas as diverse as graphic design, calligraphy, printing, geometry, mathematics and engineering.
Before the war
Before the Font Wars there were two markets for printing: The printing Industry, such as newspapers, magazines and books. And the rest of the market, such as offices, small businesses and educational organizations. The printing industry used a process called phototypesetting, which was used between the mid 20th century until the 1980’s, making it a bridge technology between letterpress and desktop publishing (Hermann, 2016). Designers who used phototypesetting would use a computer terminal to set the type that would be used for headlines, then this type would be projected in paper which would be manually placed in position and then photographed to create a printing plate (Lemon, 2017).
For the rest of the market, the easiest way to print a document was using a dot matrix printer, which used a simple grid to print bitmap characters. This printing process required that the printer was equipped with an expensive hard disk and cartridge. Dot matrix printers also had a very restricted range of fonts (Computerphile, 2014). These characteristics meant that dot matrix printers were expensive and inflexible, so they were mostly used in places which didn’t require a diversity of fonts.
The desktop printing revolution was possible because of the conjunction of 4 technologies which launched very close in time: The 1984’s Macintosh which had a graphical user interface which popularized “what you see is what you get”. The Apple Laser Printer, launched also in 1984. PageMaker, a layout and publishing software launched in 1985 for the Macintosh and in 1986 for IBM PC (Crocker, 2019). And finally PostScript, a revolutionary technology created by Adobe, which enabled scalable fonts.
PostScript was developed by Adobe founders Warnock and Geschke, who had worked together at Xerox PARC in the 70’s. While working at Xerox they encountered the problem of how to display and print fonts. Bigelow (2020) argues that at the center of the Font Wars there was a fundamental question: What is the best way to turn traditional printed letter forms into digital fonts for computer screens and printers?.
At the time there was a discrepancy between screen resolution and printer resolution, screens had a resolution of around 72 points per inch, whereas the printer resolution was around 300 points per inch. This was solved in the 1970’s by creating bitmaps of each character, size and font, which were then stored in a file (Warnock, 2018). Bitmap characters had two advantages; they were high quality and were fast to display. On the other hand, they were inflexible as they could only be used for the size they were designed for. The other disadvantage is data size, Eliyezer Kohen (Computer History Museum, 2016) explains that a 600 dpi font with all the characters would require 100Kb of space, which was very expensive at the time. The variation in screens at the time was also a problem, if fonts were created to accommodate for screen variations from 72 dpi to 2000 dpi, it would need several GB of data.
A solution for this was to use the outline of the fonts instead of a bitmap. At Xerox they used a language called JaM, in which they implemented a geometric model which used third-order Bezier curves to represent characters, then these outlines could be filled in with solid colours (Warnock, 2018). The font outlines would be designed at high resolutions and then would be scaled when displayed. Using outlines solved the problems present with bitmap fonts, now you could save one outline per character instead of having multiple bitmaps for different sizes. Outlines also afforded font designers to transform control point coordinates to create any required size or shape, which meant that characters could be rotated or edited in any way. This new technology reduced data size in exchange for computational time (Computer History Museum, 2016).
This way of using outlines wasn’t a perfect solution, it came with its own issues, for example: symmetry control, preservation of similarity between characters, scaling to very small sizes might make a text unreadable and finally controlling diagonals (Computer History Museum ,2016). In the early 80’s Adobe engineers approached this problem by using third-order bezier curves and additionally “distorted the outlines of each letter so that the vertical and horizontal boundaries would align with the pixel rows and columns” (Warnock, 2018) They also decided to make the curves to make the letters less fat, which reduced the visual weight of the letters. This new standard was called PostScript.
Warnock (2018) recounts that in 1983 Steve Jobs asked to meet with the team at Adobe. At the time Apple was working on the Macintosh and had developed a graphic interface called QuickDraw. Jobs had identified the advantages of using PostScript in the new Macintosh computer, as it was a programming language it would be easy to develop a program to convert QuickDraw to PostScript. It was this flexibility which made PostScript so popular with other
customers besides Apple. In this way Adobe created a successful business model by licensing PostScript to other companies, thus becoming the dominating standard for fonts.
The war
In 1987 Apple made a business decision to stop paying royalties for Adobe’s PostScript and Type1 fonts. Bigelow (2018) explains that Jean-Louis Gassee, then Apple’s technology vice president, opposed Adobe’s control of both technologies and sought out to create an in-house font technology. Apple hired Sampo Kaasila who had worked at Imagen, a laser printer company which developed a conic curve technology which briefly competed with PostScript. In order to avoid competition with Imagen’s conic curve technology, they decided to use quadratic B-Splines, which offered faster rasterization of quadratic curves. This new standard was called TrueType.
TrueType used additional information to avoid problems frequent with outline typefaces, these “hints” were operations which modified the scale of the coordinates before the character was filled. In this way they were able to move control points to desired positions to create characters that scaled without compromising legibility. Kohen (Computer History Museum, 2016) explains the process to create TrueType fonts: The first step was to specify the outlines of a glyph whether doing it from scratch or by using any of the digitised font outlines created by professional foundries. The second step would be to create the hints by using an interactive editor, this step is the most time intensive and requires some knowledge of both typography and engineering. The third step is to use a compiler to produce the low level TrueType code, which can be also fine tuned later. A final TrueType file includes an outline, an interpreter engine, a control value table and a graphic state.
In 1989 Apple and Microsoft announced this new standard at the Seybold Desktop Publishing conference, claiming that it was superior to PostScript (Bigelow, 2018). To ensure this new standard was competitive against Adobe’s PostScript, Apple commissioned URW —A firm which had an extensive library of digitised fonts and had created the Ikarus software for digitising, storing and processing letter forms — to create a translator from Ikarus to TrueType Outlines. In 1991 Apple released the first product version of the TrueType rasterizer in Macintosh System 7.0. Later in 1992 Microsoft released the Apple TrueType rasterizer integrated into Windows, as well as FontPack to ensure and encourage the use of the new standard.
Adobe’s Response
The first response from Adobe was to separate the Adobe Type 1 technology from PostScript, so the Adobe Type1 fonts could be used in Macintosh and Windows (Bigelow, 2018). The second response was to publish the Adobe Type 1 specifications in 1990, this was done to spark the development of new Adobe Type 1 fonts, which were already very popular with the design industry.
Peace
Throughout the 90’s both standards remained popular in different markets, TrueType in offices and home computers, and Adobe Type 1 in the publishing and design industry. Even though this seems peaceful, Hermann (2020) explains that there were major compatibility issues between font formats and operating systems, for example a TrueType font for Mac OS wasn't necessarily compatible with windows, even if both used the TrueType standard.
Bigelow (2018) explains that Apple launched TrueType GX in 1994 as an enhancement to TrueType, which fixed issues with characters outside of latin characters. Microsoft decided to not adopt this technology, instead they proposed a new technology called “TrueType Open '' which would solve the problems already ientified by Apple.
In 1996 Adobe and Microsoft worked together in the development of OpenType (.OTF), which incorporated the data structure from TrueType, the additional line layout of TrueType Open, and also supported Unicode encoding (Bigelow, 2018). OpenType solved the compatibility issues resulting from having too many standards. It took Microsoft until Windows 2000 (2000) and Apple until Mac OS X (2001) to incorporate OpenType. All standards were supported and used simultaneously for many years.
Around 2016 OpenType variable fonts were announced, these new variable fonts make it possible to have many weight variations in a single font. “This is done by defining variations within the font, which constitute a single- or multi-axis design space within which many font instances can be interpolated.” (Hudson, 2016). This new technology is especially useful in the development of web applications, where fonts have to adapt to an extensive range of devices, from desktop computers to phones, to smart watches; variable fonts offer the flexibility to adapt fonts to these new platforms.
Conclusion
We might wonder why fonts are so important as to unleash a decades-long war, with millions of dollars, time and resources spent. Designer Ellen Lupton (2010) explains that typography is a tool for shaping content, giving language a physical body and enabling the social flow of messages. Bigelow (2018) argues that fonts enrich the capabilities of operating systems and applications, all systems benefitted from the multiplicity of fonts.
The many innovations in typography that surged before, during and after the Font Wars undoubtedly benefited Apple, Adobe and Microsoft to become the dominating companies in technology in the 90’s and early 2000’s. According to Bigelow (2018) they weren’t the only ones who benefitted, digital reading has boosted literacy because of the availability of fonts outside of the standard latin characters, he explains “Billions of readers and writers in hundreds of languages now use digital font technology to communicate every day, throughout the world”. The Font Wars showed us that competition can foster innovation and technological development, at the same time they also taught us how cooperation can extend these innovations and make them accessible to everyone.
Bibliography
Bigelow, C. (2020). The Font Wars, Part 1. IEEE Annals of the History of Computing 42(1), 7-24. https://www.muse.jhu.edu/article/761998.
Bigelow, C. (2020). The Font Wars, Part 2. IEEE Annals of the History of Computing 42(1), 25-40. https://www.muse.jhu.edu/article/761999.
Computerphile (2014, January 31). The Font Magicians - Computerphile [Video]. YouTube. https://www.youtube.com/watch?v=jAdspOtgciQ
Computer History Museum (2016, August 30). TrueType: The Digital Font Technology, lecture by Eliyezer Kohen [Video]. YouTube. https://www.youtube.com/watch?v=nWs65Uejd4M
Crocker, S. (2019). Paul Brainerd, Aldus Corporation, and the Desktop Publishing Revolution. IEEE Annals of the History of Computing 41(3), 35-41. https://www.muse.jhu.edu/article/734469.
Hermann, R. (Typography Guru) (2016, April 18). Phototypesetting with the Berthold ‹diatype› [Video]. YouTube. https://www.youtube.com/watch?v=76qwCF6ThLs
Hermann, R. (Typography Guru) (2020, March 9). The Font Wars—PostScript, TrueType, the Mac and the success of desktop publishing [Video]. YouTube. https://www.youtube.com/watch?v=5X9Dj7tBlkg
Hudson, J. (2016, April 4). Introducing opentype variable fonts. Medium. Retrieved November 15, 2021, from https://medium.com/variable-fonts/https-medium-com-tiro-introducing-opentype-variable-fonts-12ba6cd2369.
Lemon, D. (2017, January 27). The font wars. pastemagazine.com. Retrieved November 17, 2021, from https://www.pastemagazine.com/design/adobe/the-font-wars/.
Lupton. (2010). Thinking with type : a critical guide for designers, writers, editors, & students (2nd rev. and expanded ed.). Princeton Architectural Press.
Warnock, J.E. (2018). The Origins of PostScript. IEEE Annals of the History of Computing 40(3), 68-76. doi:10.1353/ahc.2018.0025.