Skip to main content

A report from the Grinnell 2017 letterpress workshop

This musing serves as my report from yet another summer workshop.

The week of 31 July to 4 August 2017, I participated in the Summer 2017 Grinnell Letterpress Workshop led by Jeremy Chen and held as part of his Innovation Fund letterpress project. While this workshop was very different than the traditional Grinnell workshop, which focuses primarily on discussion, I very much appreciated the consideration of skill building in letterpress and the ability to reflect on the role of the College’s new letterpress in the curriculum.

I have divided my report into three parts. I begin with a narrative of my experience in the workshop. I continue with some reflections on the role the letterpress might serve in one or more of our courses. I conclude with some notes on the value of hand-printed output as compared to the now much more common laser printer output.


I entered this workshop with somewhat different experience and different perspectives than my fellow workshop participants. I have taken three of Grinnell’s studio art courses, so I was able to approach the workshop informed by the design and problem-solving considerations that those courses built. I participated in the letterpress workshop of two summers ago, which informed the way I was thinking about the use of letterpress. I am also a computer scientist, which affects the way I think about tools and processes.

That last characteristic was perhaps most evident to the larger group when we were talking about setting up the inkjet printer to make negatives; I quickly suggested that rather than running through a series of new settings each time, we save the group of settings as a printer option. But it was also obvious (at least to myself) in my approach to the project. I had started the project using Adobe Illustrator, but I quickly switched to LaTeX because I think better about typesetting in LaTex. Using LaTeX both freed and constrained me.

I will admit that I was a bit disappointed to hear on the first day that we were limited to only one color in our project [1]. I was also disappointed to hear that it was difficult to use typefaces smaller than about 36 pts (about 1/2" x height). Those two issues meant that I could not attempt the project that I had considered in the previous workshop and which I will write about in the next session.

Instead, I ended up considering what I am calling a MathLANifesto, a statement of principles for members of the Grinnell CS Department. The name riffs on the term MathLAN, which we use for our network of GNU/Linux workstations. In thinking about the MathLAN, particularly as we change our model of system administration, I had drafted a longer MathLANifesto and I thought it would be interesting to see how it might translate to the letterpress.

My initial plan was to create a short booklet of principles. It took me a bit of time to find an appropriate set of instructions for laying out an eight-page booklet [2]. I enjoyed thinking about how to turn the longer principles into short phrases and considering what image might go with each phrase. But once I started thinking more about fonts, I realized that computer scientists use LaTeX and I should therefore plan to use Computer Modern as the font and perhaps even do the layout in LaTeX.

But I had difficulty finding a good eight-page booklet layout in LaTeX. It also wasn’t clear to me that the 12x16 size or the grade of paper we were using would make good booklets. And so I switched to a broadside layout. Unfortunately, I don’t do a lot with font sizes in LaTeX, so I ended up doing a 9pt 3x4 inch document and then scaling it up by a factor of four.

Turning my negative into a printing plate also went less well than expected [3]. Because we needed to print and align three negatives, I ended up losing a bit of each letter due to not-quite-perfect alignment. In the future, I would likely pay the extra to send off and have the negative generated by professionals [4,5,6].

Printing went more-or-less as I expected. Slower at times, faster at others. It was useful to think carefully about the various issues involved. And I kind of like the way the incomplete letters look. A print specialist would hate it, but I like the way it looks like we’ve been doing this long enough that the letters are starting to rub off.

Were I to do this again, I think I might combine the two approaches I considered. I would do a one-page broadsheet, rather than a booklet, but it would be composed of small blocks, each of which contains a part of the work and an accompanying image. Such an approach might best reflect the strengths of letterpress.

The letterpress in CS

In the 2015 letterpress workshop, my primary idea was to have the students design annotated algorithms in CSC 301, our upper level Algorithm Analysis class. The model was that we would start with an algorithm at the center of the page and add an assortment of annotations - dead links [7] to related annotations, explanations of tricky parts, notes on alternate strategies, example run-throughs, and so on and so forth. At the time, I had thought we might have each group then print their own annotated algorithm and have a show of sorts in the CS hallway.

But as I look at the complexity of letterpress, I think it makes more sense to have one collaboratively annotated document that we make and share with all of the students in class. I’d really like to play with how to represent the different kinds of annotations. Color is one approach, but should not be the only approach we use, since it is less accessible for those who have limited color vision. Perhaps face or some iconography would help. In any case, my plan would be to have students annotate a variety of algorithms in small groups, have a non-letterpress show of examples, and then work together on the collaboratively annotated document.

To provide some individuality to the work, we might create a larger set of annotations than are possible to fit on one sheet and print each annotation as a separate block. Each student could then select his/her/zir preferred annotations and lay them out on the press. Since block setup/takedown is one of the faster processes, that method would allow students to have the sense of individual ownership that comes from making something that is yours alone along with the shared community that comes from collaboratively building the set of annotations. We could give students a bit more freedom by using two or three colors, which would also encourage them to think about how to group the annotations.

I could see some value in having my students collaborate with Chen’s, if he were amenable. I know that my CS students learn a lot when they work with a client who does not really understand our discipline and Chen’s students might find similar value. Making work for others is different than making work for yourself, but it can still both be a valuable learning experience and provide a sense of ownership and creativity.

Letterpress vs. Laser printing

Couldn’t we just do this with a laser printer? Yes, but I don’t think the sense of accomplishment is nearly the same. It feels much different to hold something you have just drawn off of the letterpress than it feels to hold something that has come off of a laser printer. In the former case, you feel much more sense of ownership, even though the content is the same. In effect, you own not only the content, but also the process for displaying the content. To put it another way, a work printed on the laser printer feels ephemeral and disposable while a work printed on the letterpress feels archival and perhaps even a mixed form–not just a presentation of content, but a work of art or craft.

I consider, for example, the famous map of Napoleon’s march on and retreat from Moscow. Part of the strength of the piece is that it feels crafted, and not just generated on a high-volume printer. Of course, this piece would be crafted even if it was printed on such a printed, but the feel of how one overlays the advance and retreat so clearly reflects printing technology that it’s impossible not to consider the letterpress (or other printmaking) technology.

Since I am now teaching data science and data visualization as part of CSC 151, I may also ask the students to reflect on the different impact of hand-printed infographics.

The letterpress and 3D printing

I will admit that I am intrigued by the idea of printing the type block on one of the many 3D printers now floating around campus. While I don’t think high resolution is possible, it strikes me that we may be able to achieve some interesting effects, particularly as we use some substance to smooth out the slightly jagged 3D-printed materials. This is certainly a subject for further exploration. However, it will likely have to wait until either winter break or my next sabbatical.

So that I don’t forget, here are some links to possibly interesting resources [8,9].

Appendix: New draft MathLANifesto

Please keep in mind that I generated this under time pressure and thought of it as a draft. I must admit that there is at least one typo and one layout issue in what I printed [10].


Principles for Grinnell College Computer Science students, faculty, and staff

As citizens and computer scientists we are obligated to assess and understand the impacts of computing on society [11].

We reflect regularly on our own moral and ethical responsibilities.

We learn through open exploration and inquiry.

We develop software using multiple languages and libraries.

We rely on free and open-source software, particularly GNU/Linux, to support these endeavors.

We seek projects with broader impact.

We collaborate—locally and remotely, synchronously and asynchronously, in pairs and in larger groups.

We welcome all who are interested in our field and seek to broaden participation.

We release our own work—including code and curricula—freely and openly.

[1] I understand why that limitation is necessary for this workshop. It just wasn’t what I was expecting.

[2] It turns out that I should have done a Web search for eight-page ’zine rather than eight-page letterpress pamphlet.

[3] Well, it appears it went about as well as Chen expected.

[4] I think we used about $10.00–$2000 in supplies for the negatives, as well as a bunch of Chen’s time. Having the done professionally would probably be about $40.00, but would have given me much more freedom in what I did with the work.

[5] When I say, I would pay, I really mean I would have the College pay.

[6] I think it’s worth knowing how to make the plate and to print with it. There’s also a pleasure to printing.

[7] The link on a typical Web page is called a live link because you can click on it and immediately see the referent. In contrast, dead links tell you where something is, but don’t bring you to it.

[8] Don’t you just love URLs that have no accompanying explanation?

[9] I didn’t include it above, but there was a great discussion I found in which someone said, in effect, that it was offensive to use the term printer in association with a 3D printer. What those machines do may be useful, but it sure ain’t printing. Let’s see, it’s somewhere in this discussion.

oelanich, its not a gripe it is an affront. Just because you can invent a machine that electronically/digitally ejaculates a polymer and you excite those polymers with light,sound or electricity and replicate an object does not constitute printing. Sure you can call it printing but it isn’t.We live in a country that guarantees freedom of speech but not necessarly freedom from ignorance. Evidently a guy named Chuck Hull coined the term stereolithography, without any explanation at all. The word he coined translates to three dimensional stone writing?, and that corelates to replicating something out of plastic?. Without a doubt Chuck Hull is a smart man,clearly he has no sense of etymology. [james bourland on 14 Jul 11 (16:23)]

[10] Working under time pressure is not good for me.

[11] My department chair did not appreciate my inability to create a parallel structure to the following sentences.

version 1.0.2 of 2017-08-03.