What “counts” as computer science?

image by Kevin Ku via Pexels

This post touches in part on two papers out of my research lab being presented this week at SIGCSE 2021: “Integrating Ethics into Introductory Programming Classes” and “‘You don’t do your hobby as a job’: Stereotypes of Computational Labor and their Implications for CS Education.” Post hoc, I realize that something at the heart of both of these projects is the potential benefits of expanding our notion of what “counts” as computer science.

As reported by Dianne Martin in a 1997 paper, a computer science department chair wrote a letter to the editor of Communications of the ACM in response to a proposal that social and ethical impact be integrated as an explicit strand into CS education in 1996:

… the most glaring problem is that proposed subject matter is not computer science…the content of the ‘strand’ has no algorithms, no data structures, no mathematical analysis, neither software development nor software design, no computer science theory. In short the content is devoid of every standard element present in computer science research and education. … It’s hard to imagine a computer scientist teaching these things. … Ethical and social concerns may be important, but as debating the morality of nuclear weapons is not doing physics, discussing the social and ethical impact of computing is not doing computer science.

The idea that ethics is not “doing computer science” persists over two decades later. We see evidence of this attitude every time we hear some variation of “I’m just the engineer” from a tech professional when questioned about ethics, and when ethics is barely covered at all in computer science curriculum. (Which yes, is still the case in many departments despite ABET requirements — I hear from college students all the time who say they’re not hearing about ethics at all in their CS program.)

A couple of years ago I suggested (“What Our Tech Ethics Crisis Says About the State of Computer Science Education”) that if we want to shift computing culture such that thinking about the ethical impact of computing is part of the role of every technologist, we have to change the way we teach computer science — not as a specialization or an add-on, but integrated into technical classes throughout the curriculum. Since then, the Responsible Computer Science Challenge provided funding for myself and many others to put this idea into practice. In thinking about the importance of this cultural shift, my collaborators and I decided to start at the very beginning: with introductory programming classes.

So all of this is to say, I think that if we want to thread critical ethical inquiry deep into the tech industry we need to start with education, and that means showing future technologists every step of the way that thinking about the ethical and social impact of computing is not only “doing computer science” but something that must be done in computer science. I also think that computer science departments should be hiring tenure-track faculty for whom computing ethics is their area of research. And I think that coding interviews at tech companies should include questions about ethical implications. (I have a lot of thoughts about the role of interviews but that’s a whole other post. :) )

It’s also worth noting that as we continue to grapple with the lack of diversity in computer science departments and in Silicon Valley, a huge number of the thought leaders in computing ethics are women and people of color. As someone who has degrees with “human-centered computing” and “human-computer interaction” in the title lines, and who currently writes more tweets about ethics than I write lines of code, I’ve certainly had enough people tell me that these “human” aspects are the “soft” part of CS, not “real” CS, etc. Which brings me to the connection I drew to our other paper at SIGCSE this year.

My PhD advisee Brianna Dym led an interview study with fan creators (majority women and/or queer and/or BIPOC) who do computational work — that is, making websites, modding video games, doing data science about their own communities, or even building massively successful online platforms. In my prior work on computational learning in fandom, participants also described the learning environment as “empowering” and “loving” and contrasted it to open source communities. This work motivated the current project (which is NSF-funded) — to explore how people from groups traditionally underrepresented in computing are still “doing” computing that might not look like our traditional notions of computer science.

The stories we heard from asking participants about their computational projects and how they saw them as part of computing tracked to a lot of what we already know about damaging stereotypes in computing: that computer science isn’t fun or creative, that it’s not for them. And these ideas were reinforced by experiences with CS education. A refrain we heard was: I took a computer science class, and it sucked, so this awesome thing I’m doing can’t be that. For example, one participant described dropping out of a computing camp because she felt condescended, talked down, and mansplained to; she contrasted this to learning things in fandom, which felt “empowering.” And though sometimes this alternate learning space led to participants pursuing careers in tech, other participants expressed not being “good” with computing even as they talked through the computational skills they learned in fandom. In undervaluing their skills (just modding video games, just writing HTML, it’s not “real” code), someone might not see computing as something they had already started and might continue to pursue. Overall, our findings suggest that it is important (e.g., in the computer science classroom) to break down not only what a computer scientist is but also what counts as computer science — particularly in terms of encouraging participation from traditionally underrepresented groups. And also, to potentially meet them where they are.

And as one participant put it: “If you want to broaden your pool for computing, you have to look in unconventional spaces. If you stay in the conventional space, you’re going to get a lot of white dudes.”

Or, as my favorite coding drag queen on TikTok says, “Computer science professors are like, women don’t want to learn how to code. Maybe they don’t want to learn how to code from you.” In fact, I liked this sentiment so much that I build on it to explain our findings!

In sum, I think there are a lot of reasons why reducing gatekeeping in computer science would be a positive thing. And when computer science departments are considering what to teach, what kinds of research is encouraged, and who they hire, broadening that definition might also broaden the pool of participation.

Fiesler, Casey, Mikhaila Friske, Natalie Garrett, Felix Muzny, Jessie J. Smith, and Jason Zietz. “Integrating Ethics into Introductory Programming Classes.” In Proceedings of the 52nd ACM Technical Symposium on Computer Science Education (SIGCSE’21). New York, NY, USA: ACM. 2021.

Dym, Brianna, Namita Pasupuleti, Cole Rockwood, and Casey Fiesler. “‘You don’t do your hobby as a job’: Stereotypes of Computational Labor and their Implications for CS Education.” In Proceedings of the 52nd ACM Technical Symposium on Computer Science Education, pp. 823–829. 2021.