12 Things I Learned in Type School

This essay is adapted from a process book summarizing my year as a student in the Type@CooperWest Extended program. It’s a collection of thoughts, projects, and sketches. I’ve learned a lot of lessons, both big and small, over three terms of studio and history classes, guest lectures, and workshops.

In no particular order, here are twelve things I’ve learned about the practice of type design, the world of type, and the creative process.

[ 1 ] My love for type

For as long as I can remember, I’ve loved type. As a kid, I’d beg my parents to buy fonts for me that I could load up onto my Mac Plus computer, Aldus PageMaker, and print out on my ImageWriter. In high school, I had a subscription to Font & Function magazine.

My interest in typography continued in school and work. At the MIT Media Lab, I experimented with writing code to draw letterforms that could be animated in 3D. Working in motion graphics, and later in product design, I’ve always been interested in how to use type to communicate and express ideas.

But when I was considering applying for the Type@Cooper West program, I wasn’t sure if I’d enjoy actually designing type. Isn’t it just a lot of adjusting vector points around on the screen? I thought it might be too tiring and ruin what I liked about type. But I had nothing to worry about. I’ve been surprised this year by just how much I love working on type. Sure, vectors and béziers can get frustrating at times, but when you get in the zone, it’s meditative and addictive.

I learned a lot about drawing bézier curves themselves. When you’re tracing a drawing in Robofont, you want to add the extreme points first with straight segments first, then “pull out” the curves that connect them, then make them tangent or continuous points as needed. In general, when you have connecting curve points, you want the slope on either side of the curve to be as close as possible to create a smooth transition.

It feels a little like a hack, but the Speedpunk Robofont extension by Yanone can be super helpful in guiding your eye on where your curves are stuck.

I love that when you’re working on letterforms, you’re in a zoomed-in 1000 by 1000 pixel grid. You don’t have floating point numbers (as far as I know), which makes drawing straight, 90-degree segments a lot simpler and lets you have, say, uniform overshoots across your character set. If you had to worry about decimals when drawing, it could make you batty.

We learned a few tips and techniques for drawing letters. Curved glyphs like s and 8 benefit from drawing them backwards and upside down.

In the first few weeks we also learned about the translation and expansion contrast models. It was so interesting to learn that the rules that define type design, whether cast in metal or vectorized on screen, derive so much from two different types of tools for writing: the pointed and broad nib pens.

[ 2 ] Positive and negative space

Type design is about both the positive and negative spaces — spaces between letters and inside of them. In a regular weight, a word like minimum should have evenly spaced vertical stems, and the white spaces between them should have the same width. A best practice is to space your characters as you go. You want to spend time in Space Center, deciding the left and right side bearings between straight and curved control characters. In a typeface, usually n and o should be the same width visually.

[ 3 ] Digital revivals

A lot of type from history hasn’t been made digital (yet). You make a lot of decisions when digitizing a typeface. Even if a typeface already exists in a digital version, you might be able to revive it better — updated for the present day, or truer to the original. In a lecture at the SF Public Library in February, Loïc Sander introduced us to his Didot-inspired Trianon family, a lovely reimagining of the Didot family style.

In the fall term, my project involved digitizing Farmer Old Style №7, a typeface from A. D. Farmer & Son Type Founding Co., which we found in a printed type specimen from 1899. I called my revival typeface Farsevo.

[ 4 ] Italic slant angles

These days modern text italic styles have a slant of about 8 to 10 degrees. But in earlier times, they were sometimes more extreme. In the break between our first two terms, I worked on reviving the italic of Farmer Old Style №7. I dialed the extreme angle back a bit, but I think it could still come back a few more degrees.

[ 5 ] Lettering

I came into this program not knowing much about lettering, and not even realizing that it’s a thing: a practice and artform with its own interesting history. These days calligraphy and sign painting have their fans and are probably still gaining in popularity. You can watch videos of masters showing off their lettering practice on Instagram. My classmate Kel Troughton is big into lettering and researching historical sources, and his interest inspired me to learn more.

In the break between spring and summer, I worked on a typeface based on show card lettering from the 1918 manual Lettering for Commercial Purposes by William Hugh Gordon. It was fun trying to regularize the lettering and make the shapes suitable for a font.

The summer workshop with Erik Marinovich and James Edmondson was extremely fun, but day two, when James saw what we had worked on in day one, was a shocker. (“What the heck were you all doing?” we could hear him thinking.) A type designer has to think about systems and readability, not just inventing novel forms.

[ 6 ] Finding inspiration

Inspiration can come in unexpected places. In the fall term, we had a weekend workshop on script typefaces with Richard Lipton. We began with drawing scripts in different sizes at 45-degree angles, then worked on how to bring them into the computer and turn them into functioning typefaces. Richard was encouraging us to come up with novel ideas for script faces. As I was drawing and starting to fill in my script letter outlines, he said, “You should stop right there!” It led me down a different path, an idea for a script with an inline fill.

We had a couple other workshops, namely with Erik and with James earlier in the year, where we were encouraged to try different tools, different media, different ways of creating letterform shapes, working fast, and changing up our process.

[ 7 ] Going wide

In our present-day visual culture, it seems that wide lettering is really only used in one place consistently, in chrome decals on the backs of cars. You’ll see it in a few other applications, like in signs or logos on appliances, but not many other places. Not sure why exactly, but I think it’s because wide fonts are not very efficient in how much space they use to communicate information.

My spring term was spent making a wide, bold sans serif face in a roman and italic style. I landed on the wide proportions because I was interested in an inline stroke idea, and Tânia suggested focusing on refining a filled face that could be a good foundation for the decorative ideas I could pursue in the summer term.

[ 8 ] The importance of counter shapes

Counters are the shapes inside letters, both closed shapes like e and o, but also the concave regions in m and n. It follows from what we learned about positive and negative space that counters are important, but their importance in establishing a visual system extends beyond a single font style. In a workshop in June, Andy Clymer explained how punch cuts were tapered, and the same punch could be used for a larger size by cutting deeper into the lead. When designing different weights of a family, you should think like a punchcutter and preserve the counter shapes so the letterforms feel related.

Sasha Tochilovsky also brought up a related idea in his workshop in March. When you’re pairing different fonts to use in a design, you can compare the counter shapes in the letter a to be a signal of whether the faces will work well together.

[ 9 ] Getting thrown for a loop

In the first class of the final term, Tânia was detailing all the things that were due at the end: three completed font masters, a process book, samples of your typeface in use, a poster, possibly a website. And then the teachers threw us for a loop. They said, “Put your projects aside for a week. You’re going to work on a new type project based on a sample we’re giving you.” They had gone through Frank’s library of photos of street and shop signage from across Europe and picked out a different one for each student to turn into a font. The class was surprised and a bit stressed, but we got to it and showed our work the next week.

I started with the lettering for ‘Gonzalez’ and worked out a set of lowercase letters that could work as a connected script face.

I’d say getting thrown for a loop can be best thing. Each of us got to practice something totally different from what we had been heads-down on, and more than that, we got to see just how much we’ve learned. I was impressed by how great the one-week projects were, and they were miles from what we could’ve accomplished just a few months before.

[ 10 ] Not business school

For my final term project I wanted to turn my second term project into a more useful and usable regular-width face. I asked the teachers what they thought about doing a type family with four masters, so I’d have a range of weights for both roman and italic styles. Frank’s advice was that four masters was too predictable and encouraged me to pursue more ideas, not just getting “caught in the loop of endlessly refining.”

My final project was a sans serif family called Peasy.

I really enjoyed the last few weeks of class. I dug into the inline stroke idea I started with in December and developed it into a stencil typeface that I think is more expressive, interesting, and impactful than my initial sketches. Drawing ideas on paper helps a lot, and so is the prodding of good instructors. Frank said at one point in class, “This is art school, not business school.” It stuck with me.

[ 11 ] Where to go from here

I attended Typographics, a type conference held in New York City in June. The talks were inspiring, and it was fun to meet some of the students currently enrolled in the Type@Cooper(East) program to take notes on the two programs. Some of my biggest takeaways came from a lunchtime panel discussion geared toward type designers who are starting out.

The panelists made the point that type isn’t actually done until someone uses it, and that type design shared on social media is really just for other type designers, not for the people (e.g. graphic designers) who will complete the loop by using your type. Also, not all type designers are necessarily good at graphic design, and it’s ok to let graphic designers find the best way to showcase your faces in type specimens and marketing materials — you don’t have to do it all yourself.

Finally, they offered the suggestion to share your typefaces in progress with friends to try out in their designs. Other designers can give you advice on what they see in your fonts, what they might be good for, and what you should change.

[ 12 ] The long tail

When you tell non-designer friends you’re working on a font, they’re probably picturing that you’re making twenty-six letters. If they think for a moment, they might remember there’s uppercase as well as lowercase, and they look different, so that makes 52. If they’re really smart, they’ll remember numbers, and punctuation. This term it really sunk in that type design is a long tail effort. The further you get, the deeper you dig, and the more you realize is still there to find. Your typeface will never be complete.

When it comes to character sets, you start out by focusing on a few lowercase glyphs (e.g. hamburgefonts) because those shapes help establish rules you can use across the remaining lowercase letters. And lowercase a–z are the most important to get right because they are the characters that make up the vast majority of type when someone uses your font. Uppercase is important, too, and you should consider all-caps applications. Punctuation like the period, comma, question mark, exclamation point — they’re all essential. Numbers: do you want proportional lining figures, oldstyle, tabular variants? Dashes, parentheses, brackets, and more symbols follow. What about non-English settings of your font? You’ll need diacritics for your lowercase and uppercase characters, as well as characters like ß. Some common ligatures would be helpful for letters that tend to collide, and how about small caps? Borders, ornaments, and dingbats can add a lot of personality.

If you’re really serious, you’ll want to consider extending your character set to accommodate Cyrillic, Vietnamese, Greek, and possibly Arabic, Chinese, Japanese, or Korean.

The long tail refers not just to which glyphs you include in your typeface. It takes time to develop a family of typefaces that work well together. It also is a long process to “master” a font and prepare it for release with steps like kerning, hinting, establishing OpenType settings, as well as extensive testing and proofing.

It can take a long time to release a typeface. It also takes time for your eyes to develop. When you come back to a project after a while, you’ll see new things to try that you hadn’t noticed before. You see more and more things as time goes on.

Type design is a long journey, but it’s fun!


Much appreciation, respect, and love to Tânia Raposo, Frank Grießhammer, James Edmondson, and Grendl Löfkvist for being enthusiastic and supportive instructors.

The Type@Cooper West program ended its run in 2018, but if you’re interested in learning type in San Francisco, I highly recommend the Type West program at the Letterform Archive.

Check out my type work and experiments at Typotopo.com and follow typotopo on Instagram for the latest.