Saturday, August 23, 2014

Introduction: An overview of libre typography

This blog

     In this blog, I'm going to be writing about designing a new typeface, and releasing it as an open source font. Designing a high quality typeface takes a long time—months, often years—so this blog will document my progress along the way. The open source program FontForge will be used to make the font. Unlike other font projects, I'm also going to write about the design process for the glyphs, in a letter-by-letter way, so that more people can learn about type design. And since each creative decision will be documented here, I hope that this will be the first open-design font. A person should be able to follow the process, and at each design junction, they should be able to fork the design itself and go in a new direction to make a different typeface.

An overview of libre typography

     This chart shows a few representatives of some of the different kinds of serif type that exist (slab serifs like Rockwell not included). The fonts are grouped and sorted by their type classification, and color coded by their license.

A crash course in type classification

     On the left column are old styles, transitionals, and didones, sorted from most "old style" to most didone. The differences between them are pretty well documented. The fonts at the top are the oldest old style types*—they're usually used to evoke a sense of the colonial era. Transitionals are a really broad group that can be understood to encompass anything between an old style and a didone. The most common example, though probably not the most representative, is Times New Roman. Didones or modern serifs are the kinds like Bodoni. With almost perfectly vertical weight stress, and hairline serifs, they are often associated with high fashion and high mathematics. Some designers break these groupings down into even finer categories, but the names don't matter, only that you know the differences between the two ends of this style spectrum. The most readable fonts tend to be found in the middle of this spectrum.

     The right column makes up the type families that don't fit well into the whole old style–transitional–didone spectrum. In the middle are the claredons, like Century Schoolbook and Charter. They are a lot like didone typefaces, though with much less stroke contrast. They also tend to be heavier, more squarish, and sometimes wider than a didone. A similar group doesn't really have a name, though here I'll refer to them as "computer claredons". They were optimized for the screen and other low resolution media, so they tend to have short square serifs and squarish shapes to fit pixel grids. A few of them approach slab serif-ness.

     The handful of fonts in the upper right corner don't fit well into any category, though they could be lumped in with the transitionals. They have old style characteristics, but are more angular and have simpler serifs and terminals. I'm not aware of any name for them (though people keep coming up with different ways of classifying type), but it includes some notable fonts like Proforma and the (in)famous Palatino. Gentium is a strange font that belongs in a category of its own, but I included it on the chart since its one of few open source fonts out there.

Font freedom

     Another way you can group typefaces is by their license, or how much freedom fonts have. Font freedom is important because it's about how the font was created, along with what you're allowed to do with it. Font freedom also tends to correlate with comprehensiveness—commercial fonts are less likely to support foreign writing systems and phonetic glyphs that their free counterparts.

   The fonts in pink are true free typefaces. Some, like Linux Libertine and Gentium were open source collaborative projects, others like Merriweather and Crimson were gifts from individual designers who released the end result of their work with a free source license. The fonts in coral are also technically libre fonts, though they were either gifts from corporate foundries, or were made by corporate foundries on commission by the open source community. Many of these fonts are subset gifts, and the foundries often sell premium versions of the fonts which include more glyphs and features. This can lead to some confusion, as the glyphs added by the foundries themselves are not always the same as the ones that open source volunteers may have added to keep up, even though the two versions share the same or similar names. DejaVu serif is in the middle—it was made by a corporate foundry, but has since been seriously expanded by volunteers.

     The gold fonts are of mixed status. No one owns these typefaces anymore, so anyone can make their own versions of these typefaces. These fonts have had their patents (or copyright, outside of the US) expire, so open source designers were able to copy them and give out free versions. There are now for example, at least three open source clones of Times New Roman out there. Bodoni and Century Schoolbook are in a slightly darker shade because even though it's probably legal to, as far as I know nobody has made a serious effort to digitize them as free fonts.

     The dark gold fonts are what people in the FOSS community call "free-as-in-beer". The fonts, though not necessarily the typeface designs, are not free to modify or distribute, but they come with so many operating systems and design apps that almost everyone has a license to use them. Unless you're on a FOSS platform like Ubuntu. Their licenses are heavily diluted but technically still proprietary.

     The fonts in gray and black are commercial fonts. They are not publicly available (though somewhat easy to pirate), and designers usually have to seek them out and pay hundreds of dollars, sometimes thousands, for a full font family, to use them. While this chart doesn't show it, there are hundreds of high quality professional fonts in every category that a handful of corporate foundries sell. While most of them are decorative typefaces for book covers and movie posters, they include most of the fonts that magazines and textbooks use. The chart above focuses on libre typography, so professional fonts are heavily under-represented, but for every free font, there are several commercial fonts that it was meant to replace. The fonts in black—Minion and Warnock—are distinguished because at the time of writing, they are protected by patent (the last Minion patent will expire in late 2018). While it is legal (though perhaps not ethical) for an American designer to clone—trace but not outright copy and paste—typefaces like Proforma (which is old enough that it would have lapsed into the American public domain by now anyway), anyone who tries to clone Minion is breaking the law (though there are serious questions to be raised aver the validity of the Minion patent).

Let me be clear. There’s nothing wrong with commercial fonts. There is a great deal of hatred within the FOSS community towards companies such as Microsoft, which they have come to style as “M$” to villify its commercial nature (sometimes for good reason—Microsoft has played some dirty tricks against free software in the past). Fortunately font DRM never really caught on, and so type design has been spared much of the hate directed at the copyright industry. But there does seem to be a disproportionate amount of hate going the other direction. Type designers deserve to make money from their work. I’m a type designer. I like money. But that doesn’t mean it’s wrong to create free replacements for commercial fonts, any more than it is to create free software to replace commercial software.

Just because we already have a font doesn't mean we shouldn't make more

      The conventional wisdom in the free and open source movement is that it's better to contribute to existing projects and make them better than to go off and start a new one. That makes sense for software. It's better to have one good open source word processor than fifteen half-finished word processor code bases. But typefaces are not apps. One key difference is that typefaces reach maturity much faster than apps do, if ever. Once the design of the glyphs is finalized, a typeface is largely complete. Building the font involves filling out all the special characters, hinting, and kerning, and once that's done, a font can be called finished. Many of the major open source fonts are 'finished'—they have all their Latin, Cyrillic, and Greek glyphs, and the only work to be done after that is perhaps to add ligatures and port them over to newer formats. While for most software, you can always find an area to improve, or a feature to add, with fonts you can to some extent finish them and put them on a shelf. But just because Linux Libertine is done doesn't mean that open source typography as a whole is done.

     There are now dozens and dozens of high quality free fonts out there, but the vast majority, like Ubuntu and Open Sans, are sans serifs. And most of the free serif fonts are found in that lower right corner of the chart, the computer claredons. It's natural for open source developers and designers to focus on their natural medium, but it's time for free typography to move beyond that and make more forays into other styles. Some free designers have made revivals of old-old style typefaces like Goudy and Garamond, but those too are also clustered near the other end of the spectrum. At the time of writing, only the Linux Libertine fonts and perhaps a few other less well known typefaces like Crimson and Poly can claim to be a comprehensive open source typeface not in one of those clusters. Don't get me wrong, there's quite a few free fonts like them, but very few of them approach the glyph coverage of Libertine.

     We have at the moment only about half a dozen to a dozen good free serif families, often with one or no representatives in each category of serif type. When you first install Ubuntu and open a word processor, there are only a few dozen fonts for you to choose from. Most of them are language fonts, meant to cover scripts like Korean or Arabic. There is not a lot of choice out there for any particular language/script. Some of this is Ubuntu's fault—there are several gems out there that the distro doesn't ship by default, and it takes some hunting for a user to find them. But free and open source typography needs more diversity—and unlike software, creating new high quality font families should be seen as anything but redundancy.

* Actually, if we go literally chronologically, there are at least two crude groupings that come before—the medieval blackletter (you all know what it looks like), and the Roman monumental capitals (like Trajan). Before that, we really start to stretch the meaning of "serif". These are incredibly crude groupings, but this isn't an article on typeface evolution.

Microsoft lets you install (but not modify) Georgia and its friends through the ms-core-fonts package.