The Markant typeface family
A short history of news typography
On the design of Markant
Typographical features
Script & language support
Markant on the screen
About the designer

The design of what eventually turned out to become the typeface family Markant was inspired by my fascination of typefaces for newspaper printing. Especially the typefaces designed for body copy use – the long running text, in narrow columns, with little or no leading – I found to be a worthy subject for exploration.
In the history of newspaper printing, the design of newspapers have moved from it’s earliest book typography roots, across changing formats and printing technologies, which have all dictated new typefaces to be designed especially for these specific generational instances. Since the beginning of the 19th century, newspaper typography can be said to have come into it’s own, with set of conventions that are still with us today.

Several technical challenges in the last century have produced works of type design that have become hugely influential within the field, an sometimes (perhaps by accident) even outside of it.

In 1922 Chauncey H. Griffith was promoted to Vice President of Typographic Development at Mergenthaler Linotype. He immediately started the development of new typefaces to replace the prevailing modern1 style faces. The issue troubling the moderns was their high contrast design. Especially the hairline parts of the cast lines could break of while printing, and counters could clog with ink and pulp. Faster printing meant transferring the cast lines with the stereotype process to a letterpress cylinder for high-speed rotary printing on endless rolls of paper stock. C. H. Griffith’s new approach was to engineer new typefaces to the printing method2. That meant drawing inspiration from the Egyptienne3 style as seen in the Clarendon4 typeface, with it’s very sturdy lower contrast design, and Theodore Low De Vinne and Linn Boyd Benton’s Century Roman, which possessed elegance and legibility. The first product of these efforts was Ionic No. 5. It was an instant success, within eighteen months it was used by more than 3000 newspapers all over the world5. C. H. Griffith and Mergenthaler Linotype continued to refine the design in subsequent iterations: Excelsior (1931), Paragon (1935), Opticon (1935), Corona (1941). These became known as the Legibility group.

Times New Roman commissioned by The Times of London and released by Monotype in 1933 after one year of exclusivity to The Times, was a design overseen by Stanley Morison6. This typeface drew heavily upon the Plantin face, and was allowed to have delicate lines and be semi-condensed, because it was printed in sheet feed letter press as opposed to high-speed rotation presses. This could only be done because The Times had a relatively limited daily run. As proceeding time allowed improvement in paper stock and printing technologies other newspapers adopted Times New Roman even in high-speed rotation letterpress. This design proved so popular that is to this day remains in use in some newspapers, but mostly see use in periodicals and books.

With the move to photolithography and offset printing in the 1960s, print quality took a turn for the worse. Photolithography involved many stages of transferring a layout from intermediate mediums to the final paper substrate. Printing speed increased, leading to ink smearing or ghost impressions on opposite pages if the ink was not applied thinly enough, in turn leading to greyish impression on an already greyish stock. Further, the paper stock had coarse fibres and a porous structure, leading to high ink spread. This was to inspire Dutch type designer Gerard Unger to design Swift for release in 1985. It was sturdy with large wedge serifs, large x-height and open apertures, but perhaps most importantly it was of a contemporary aesthetical language. It too some time for the design to catch on in a very conservative newspaper industry, but then it did have to replace the two major typefaces in use – Times New Roman and Excelsior – that had been the staple of newspaper body copy since the early 1930s.

Swift masterly integrated le goût hollandois7 with technical solutions and contemporary design sensibilities. Later Gerard Unger would take on the issue of economy of setting in his 1993 design Gulliver. By that time page composition had moved to electronic layout systems, and advances in printing technology and paper quality had opened up for more delicate designs.

Matthew Carter – another prolific type designer within newspaper typeface design – had also seen this opportunity for more refinement, and designed Miller8 for release in 1997. This was a Scotch Roman revival9, stylistically closer to Linotype’s Legibility group, and in a much more conservative design language than Gerard Unger’s designs. Yet it offered more typographical variety with intermediate weights, small caps, grades10, body text and display versions. These were all typographical garnishes that had been largely abandoned from newspapers since hot metal line casters were introduced. With computer layout applications, these sophistications could now be re-introduced.

During C. H. Griffith’s tenure, in 1929, William Addison Dwiggins was hired as a type-designer, and would become Mergenthaler Linotype principal artist-under-contract11 Famously his greatest legacy was the discovery of the optical phenomenon he coined the M-Formula. It was the culmination of a number of ideas he had collected and noted in letters to C. H. Griffith, but the revelatory moment came from his work in carving marionette dolls12. He discovered how angular features amplified shapes as they were seen from a distance, and he applied this spatial effect to the planar world of type design. Although he repeatedly tried to design typefaces using his idea, it was never exhausted in any published design.

His findings on the optical perceptual effects of angularity were predated by the work of French ophthalmologist Dr. Louis Émile Javal in his book Physiologie de la lecture et de l’écriture13 published in 1905. He worked on legibility in very small size text – from six down to one point, and commissioned type to be cut per his instructions by Charles Dreyfuss. A specimen of these fonts was published. The angularity in these types was taken to the extreme, with abundance of 90-degree angles and in some cases counters inked, entirely omitting the surrounding stroke.

Bauhaus faculty member Herbert Bayer created his highly modular Universal in 1925 on a commission from the Bauhaus founder Walter Gropius, to design a typeface for the school’s communications. This modular approach proved hugely influential, and come computers, the concept of modularity was inherent to the screens. Even the fonts in a dot matrix printer were built over a rather coarse grid.

For the 1985 launch of the Apple Macintosh personal computer, graphic designer Susan Kare was commissioned to create a bitmapped typeface for the innovative graphical user interface. The harsh aliased look was at that time softened by the CRT14 monitor it was displayed on.

As printing devices became generally available and display resolutions increased, the need for scalable typefaces became apparent. In the case of bitmapped fonts, any enlargement would make the pixels equally enlarged, or a much higher resolution laser printer15 would likewise need to enlarge the typefaces for print. Thus the need for resolution independent typefaces, which could be displayed and printed on any device, only limited by its native resolution.

Several technologies to solve this problem were developed, but the surviving one was Adobe’s PostScript16. With it, a typedesigner could describe glyph outlines with Bézier curves and have them rendered with fidelity on high-resolution output devices like the Linotype Linotronic 10117.

The problem was how to render these outlines on low-resolution devices like the computer displays? Without instructions on how, where and when to turn on or off specific pixels, the bare Bézier outlines would render on the grid of the screen in a mathematically logical way, but inconsistent in regards to what a typeface was supposed to look. It could be that stems were not of equal thickness, a major problem since the body copy sizes would optimally require a one pixel stem width, and should it fall to either side of this, it would mean that either the stem would disappear or double in width. Other problems included handling overshoot, making sure hairlines were rendered, that closed shapes did indeed close in the rendering, and that pixel patterns were alike in similar outlines.

This necessitated the inclusion of hints in the outline fonts. These hints were built-in instructions on how to render a font. Information on stem weights were included and horizontal zones of alignment were described for the whole font. Still, at the smallest sizes – and incidentally the most used sizes – this information was not enough, special screen representations had to be included with the outline font. These screen representations had to be hand edited for each relevant pixel size.

In 1996 Microsoft initiated the Core fonts for the Web project to offer a collection of typefaces optimized for screen use. Matthew Carter had already designed one of the two major components of the package, the sans serif design Verdana, and he was commissioned to design the serif design Georgia for the Microsoft Network18. With Verdana and Georgia he reversed the usual sequence of working, and started by designing the bitmapped design for screens. Only after this step was completed he would create suitable outlines that could be hinted to re-create his initial bitmap design. Georgia can be said to be a screen version of his earlier Scotch Roman design Miller.

As computing technology evolved colour displays became standard equipment, and computer memory and processing speeds allowed for the display of grey scale and colour images. This meant fonts could be rendered anti-aliased19, which is a rendering with multiple values of grey to indicate density, rather than just the binary aliased rendering. The TrueType20 font format offered a much more sophisticated approach to hinting, and the anti-aliasing could be controlled to better effect. I should however take a long time for the market dominant Microsoft Windows operating systems to implement anti-aliasing as a default. Only with the later ClearType21 sub-pixel rendering technology, updated system fonts that took advantage of this, LCD22 displays as the norm and with Windows Vista had screen typography moved past the aliased rendering technology.

A commonality between the outlined developments in typography is the outside driving factors in technology. The current move towards medium agnostic media means contents must be separated from presentation, and presentation must be generalized and automated as much as possible. The New York Times developed the Times Reader23, which employs an auto art director24 algorithm to lay out content on the fly, and all this in the newspaper’s recognizable Cheltenham typeface25. The enhanced span from previously only printed newspapers, to now also span screen representations sets forth some questions, namely if is it possible to design a typeface that is both well suited for printing and for screen use. Is it possible to, at the same time, brand a body copy typeface without getting in the way of technical limitations, and without disturbing legibility. Is it possible to create good readability on a screen?