An approximation of my talk for the Digital Bridges conference

Topics/tags: Talks, digital humanities, CSC 151

In a recent musing, I discussed the slides for my then-forthcoming talk about my plans for the upcoming digital-humanities-themed CSC151. But what about the words I will say? I was debating between ad-libbing, which is my normal mode of presenting, and reading a speech, which I’ve started to do in some situations. I’ve decided to write out an approximation of the talk as a way to prepare to present. Whether or not I actually read the talk will depend on a variety of issues [1].

If you’d like to follow along, you can still access the deck of Google Slides.

Good morning. I’m Sam Rebelsky. I’ll be one of the Obermann Digital Bridges Fellows this fall. Change slide. Let me start with some background.

Teresa said that we should introduce ourselves. Click to bring up a picture of SamR. As I said, I’m Sam Rebelsky. I use he-him-his pronouns [4,5]. I’ve been at Grinnell for twenty-one years. My colleagues know that that I am opinionated and outspoken. Since we are in a country in which Congress has proposed legislation that encourages high-schools to replace foreign language with coding [6], let me begin by saying that (a) computer programming is not the same as computer science and (b) among other things, we ask students to learn a foreign language so that they gain insight into cultures and people other than their own, which helps them better understand the other, to help them develop a stronger understanding of the structures of their own language, and to build their strengths in navigating ambiguity. Programming languages provide none of those benefits.

Click to bring up computer science picture. I’m a computer scientist.
Click to bring up connections to humanities pictures [7]. I’m here to talk about interactions between CS and the humanities from my perspective as a computer scientist.

As you might expect, my colleagues across campus [8] rely on the CS students to provide support for their digital humanities projects.
Students who have worked on these projects — I think particularly of Jennelle Nystrom, the Reverend Harry Baker, and Ezra Edgerton — regularly reported that working on these projects changed their view of CS. They also encouraged me to think about ways in which we could introduce these kinds of ideas earlier in the CS curriculum. Jennelle also helped me understand how computer scientists approach these problems differently than programmers.

For these reasons, and many others, I’m working on a new version of our introductory course that use digital humanities as a theme. Unfortunately, unlike the other projects you’ve heard about, mine is a future project, rather than a past project. Click button to bring up a calendar icon. I’m developing the course this fall and won’t be teaching it until spring 2019.

The design of the course is very much informed by current departmental practices [9]. Click button to bring up a picture of students pair programming. We teach the introductory course in a workshop style. Students read a few pages before class and spend the majority of class time working in randomly assigned pairs on a series of problems while the class mentors and I look over their shoulders and ask and answer questions.

Click button to bring up DrRacket logo. We use the Scheme programming language, a variant of Lisp. That won’t change. I’m happy to talk offline about the many benefits of using Scheme.

Click button to bring up an icon of a palette. And, for a bit more than the past decade, we’ve been teaching the course through the lens of image making. Matthew Kluber in Studio Art, Janet Davis in CS, and I worked to develop that course.

All three practices share one particularly important goal. Click button to bring up an image that represents diversity. All have been shown to encourage a more diverse group to consider and to persist in computer science. Maria Klawe, President of Harvey Mudd, notes in particular that themed introductory courses are one of the most successful ways to support women in the discipline.

I am developing this new course, in part, to encourage a cadre of students to consider the digital liberal arts from multiple perspectives and to develop students who can better support digital humanities projects. I also find this an exciting topic to explore. At the same time, I hope that a digital humanities version of the course will continue to grow diversity in the discipline. Click slide to bring up yet another arrow.

As these courses change who takes the first CS course and who persists in CS, we hope that we diversify who does computer science. Click button to show arrow pointing back to CS. Diversifying the discipline is a core part of our mission. Given computer technology’s great impact, the people who develop that technology should represent a wide variety of identities and perspectives. Click. And that can change the world.

Click button to bring up citations. Don’t say anything.

Click button to bring up next slide. So, what might go in an introductory computer science course with a digital humanities theme? I learned from sitting in on Erik Simpson’s course that it’s important to consider multiple aspects and that it’s better to focus on a few than to give a complete buffet

Click button to bring up structure image. We will spend some time talking about representations of data, focusing on XML and XML-like structures. These structures are particularly natural in a Scheme-based course. For example, Paul Graham notes that in many ways, the designers of XML rediscovered the power of Lisp’s s-expressions.

Click button to bring up generation image. There is power in making new things. While we will explore nonlinear writing, we will also consider ways to write documents that use code to support dynamic content.

Click button to bring up analysis image. I would be remiss if I did not include text analysis as a core topic.

We will also cover computer science topics. Click button to bring up CS image.

Click button to bring up citations.

To be perfectly honest, this is our introductory course in computer science. Hence, click button to bring up view with CS dominating, computer science is really our focus. What do I mean by that?

Among other things …. Click. Students learn to develop and analyze algorithms. Click. Students think deeply about the ways in which we organize data. Click. They consider the efficiency and efficacy of their algorithms and the way data representation affects that efficiency. Click. They think of code as language and explore issues of elegance. Click. They acknowledge and consider their ethical responsibilities. Click. The develop general code and algorithms that can be reused in other situations. Click. They reflect on ways their code may break and how they recover from such situations. Click. And they learn some complex ways of thinking, particularly recursion.

Click to bring up citations. Click to bring up slide on topic modeling. Since this is an analysis session, let’s consider the example of topic modeling. It comes up, in part, from conversations that Liz Rodriguez and I had when guest teaching Erik’s class. Many people use topic modeling as a black box, and that’s fine. But I want my students to approach it as computer scientists. I’ll begin by looking at how topic modeling can be used and emphasize that it is not intended to provide answers, only ideas and directions. Students will then read and dissect a simplified version of the LDA algorithm. That also provides an opportunity to think about how we store the data and what it means to store the data in different ways. What is gained? What is lost? I know that students only really understand the subtleties of an algorithm when the implement it. And they understand it even better when they play with the implementation. But we won’t cover everything. At least in this case, I don’t consider it important for students to delve deeply into the mathematics and statistics that show that the LDA algorithm is correct.

I have a lot to learn. I also have more than thirty paired readings and labs to write and both homework assignments and a large project to design. Hence, I look forward to the time that the Digital Bridges fellowship will provide me. And, as I said, this is future work. Things are likely to change. For example, yesterday’s conversations have gotten me to consider ways in which I might use APIs for Google maps or Twiter in the course. I look forward to conversations with some of you on how to navigate the relationships between the approaches and what I might or might not include.

Damn! Grammarly thinks that will take me nine minutes to read. I have seven. I could try to be fast. I could time myself. Or I could cut some things.

I timed myself. It’s closer to seven minutes. I should be fine. Reading it aloud also encouraged me to make a few changes.

Postscript: Although I discussed my thought processes in my musings about writing the blurb for this course and the slides for the talk, I had neither the time nor the energy to discuss the many edits I made while writing this talk.

[1] What issues? In part, it will be the custom I see people using [2]. In part, it will be my own state of mind. I know that having things written down helps me remember to cover everything I intend to cover. In part, it will be the technology available. And, unfortunately, ITS managed to shut down MathLAN and our Web server when I was working on writing this talk. It made the job much less pleasant.

[2] I will admit that I had assumed that more of the presenters would be reading their own speeches. But I saw very few do so on the first day. Of course, one of the more articulate presenters did read from prepared notes, or at least seemed to. Surprisingly, none of the ones I saw used the "animated slides that I’ve made key to my presentation [3].

[3] The legendary Erik Simpson isn’t using slides at all.

[4] I’ve been taught that it’s important to normalize the introduction of pronouns. People should not feel required to give their pronouns, but it’s important to feel comfortable doing so.

[5] No, I will not be reading these endnotes.

[6] Terrifyingly enough, one of the panelists suggested the same thing at the College level. I felt I had to reply.

[7] I should have found a bridge!

[8] And soon to be next door!

[9] If I could still edit the slides, I might temporarily hide the arrows and image connecting us to the humanities.

Version 1.0 of 2018-08-09.