Some things to think about before you exhort everyone to code

First panel: Two men stand at a chalkboard. While one man works on an equation, the second man thinks, "Wow, he sucks at math." Second panel: A man and a woman stand at a chalkboard. While the woman works at an equation, the man thinks, "Wow, women suck at math."
XKCD, "How It Works"

Oh, how I hate being the bearer of bad news. Yet I feel I have to tell you something about the frustration I’m hearing, in whispers and on the backchannel, from early-career women involved in digital humanities.

Here, there, and everywhere, we’re being told: A DHer should code! Don’t know how? Learn! The work that’s getting noticed, one can’t help but see, is code. As digital humanities winds its way into academic departments, it seems reasonable to predict that the work that will get people jobs — the work that marks a real digital humanist — will be work that shows that you can code.

And that work is overwhelmingly by men. There are some important exceptions, but the pattern is pretty clear.

In principle, I have no particular problem with getting everyone to code. I’m learning to do it myself. (And a million thank yous to those of you who are helping me.) But I wanted to talk here about why men are the ones who code, so that we can speak openly about the fact that programming knowledge is not a neutral thing, but something men will tend to have more often than women.

This matter is of no small concern to me. It is breaking my damn heart to see how many women I know have earnestly committed themselves to codeacademy because they want to be good citizens, to prove they have what it takes. These are my friends, and this is their livelihood, and this is the career we’ve chosen.

First, men — middle-class white men, to be specific — are far more likely to have been given access to a computer and encouraged to use it at a young age. I love that you learned BASIC at age ten. But please realize that this has not been the case for all of us.

Second, the “culture of code,” the inside jokes and joshing that you enjoy, may not be equally appealing to everyone who encounters it. This should be, but apparently isn’t, obvious.

But Miriam, you’re thinking, there are lots of examples of DH coders who started late and are now well-respected and proficient! This is true! And they inspire me all the time. But this is also why I wanted to talk a little bit about what it’s like for a woman to learn to program.

Should you choose to learn in a group setting, you will immediately be conspicuous. It might be hard to see why this is a problem; after all, everyone wants more women in programming. Surely people are glad you’re there. Well, that’s true, as far as it goes. But it also makes you extremely conscious of your mistakes, confusion, and skill level. You are there as a representative of every woman. If you mess up or need extra clarification, it’s because you really shouldn’t — you suspected this anyway — you shouldn’t be there in the first place. I have sat through entire workshops that were clearly pitched beyond my skill level because I just didn’t want to make us look bad. It’s more awkward when you break for lunch, because where are you supposed to sit? It’s uncomfortable. I am not known for being shy, and let me tell you: It is awkward.

But there are all these online communities where you can learn to code. There are! But if you are under the impression that online communities are any friendlier to women’s participation, then you, my friend, have not looked lately at Wikipedia.

Well, just practice! I did the work — so should you! Here is the real point I’m trying to make here: It is not about “should.” What women should do has nothing to do with it. The point is, women aren’t. And neither, for that matter, are people of color. And unless you believe (and you don’t, do you?) that some biological explanation prevents us from excelling at programming, then you must see that there is a structural problem.

So I am saying to you: If you want women and people of color in your community, if it is important to you to have a diverse discipline, you need to do something besides exhort us to code.

62 Replies to “Some things to think about before you exhort everyone to code”

  1. Great post Miriam! It is absolutely essential that we are able to see the gendered nature of the technosocial situation that is “code” and that we recognize the different access points. In thinking about whether or not markup can be feminist, it’s become clear to me that much of the DH field still operates under the belief that code and markup are neutral tools.

  2. For me this is part of a larger concern about the political moves towards exclusivity in some parts of the DH movement. It is important to delineate newness and the specificity of DH work. But at the same time drawing lines and saying DH is this or DH is that creates false binaries that have the potential to make individuals and whole communities alienated. Thinking about the history of women with relation to computers and how it impacts trends such as the mandatoriness of coding is a valuable analogue for considering how any person or group with a differing background to traditional DHers may feel when wanting to participate in these types of projects. These are the types of barriers I try to make as permeable as possible in my work with older pedagogues and wary students and is something everyone should be more aware of in their rhetoric.

  3. Great post, Miriam. I feel I need to respond, in particular, to this part, since you link to my blog post (thanks!):

    ‘Second, the “culture of code,” the inside jokes and joshing that you enjoy, may not be equally appealing to everyone who encounters it. This should be, but apparently isn’t, obvious.’

    I seriously considered adding in several paragraphs that discussed how male-centered and stereotypical much of existing code culture is. The jokes about “Real Programmers” are perfect example of how coders often engage in competition and status-seeking games that rely heavily upon oppressive ideals about what it means to be a man in contemporary society. I left those parts out because I didn’t want to digress too much from my general point that there’s a culture we need to be aware of and, as you so trenchantly and rightly do here, analyze, critique, and try to change.

    In fact, (and I think this point didn’t come across too well), much of the culture I want people to absorb are the technical best practices like self-documenting code, comments, and re-use of existing libraries and other codebases whenever possible. I even had another line that didn’t make the post about needing to have a Code Writing Center that could help people revise/refactor their code for readability.

    I obviously did not make it clear enough that I want us to approach coding and code culture as full-on humanists, not to accept uncritically the existing culture. I’m very glad to see you’ve made these points here, far more effectively than I could have.

  4. Though the macho culture of programming is real, I think “not coding” isn’t a very good solution. Women-only classes is one solution, because environment does matter, but this is an actual, beneficial skill.

    I come from an English background and the entire reason I’ve had the career that I did is because I learned to code. My friends (male and female) with my background struggle to find middle class jobs, and it’s largely been a cakewalk for me because being able to write well and code well is a rare skill combination.

    Thinking that exhorting people to code is somehow bad because of the (very real) gender divides and issues of programmer culture, however, a focus on abstraction is equally damning. Deconstruction without solutions doesn’t actually help anyone. It’s a bias I have, but I don’t think opting out of what will become an expected “second language” in many jobs is going to make the lives of women any better.

  5. Three cheers for Miriam!

    Hip Hip Hooray!
    Hip Hip Hooray!
    Hip Hip Hooray!

    Such an important conversation about both making room at the table for everyone and also questioning who is in a position to “invite” folks to the table in the first place.

  6. Miriam — Yes, I agree with your reading of the gendered issues about code. I am of the mind that DH should most of all be about cooperative work among people with different perspectives. That “code” in the computer programming sense is just one part of the emerging interdiscplinary field. That the whole binary (binary code? hah!) between “doing” and “thinking,” between “hack” and “yack” is in the end a false one, and that the best work we can do comes from listening and contributing across diverse approaches to the digital (and to the humanities). Thanks for this! — Michael

  7. I firmly believe that digital humanities has been pigeon-holed into web developing, or “coding,” and this is largely due to the power and influence that literary scholars exercise over DH. It’s possible to make extraordinary contributions to humanities disciplines using technology that doesn’t involve coding, databases, XML, etc. I believe DH must expand to inform students and scholars of non-textually-based disciplines of technologies that are relevant to their own research. Until DH begins to reach out to all humanities fields, I am reluctant to participate in the community even though my academic pursuits often involve technology and my job clearly resides in alt-ac/DH. Note that when people talk about “coding” in DH, it usually never refers to projects like this or this or this.

  8. love this, as a person who studies IRL communities intersting to see that the issues translate into the virtual world and that they are being addressed. In a prior life I co-ran a software start up but someone else did the coding while I handled the business side of the house. I’m ok with techy people because of that, but can’t imagine if I didn’t have that experience.

    I guess I do share the fear that coding moves ” towards exclusivity in some parts of the DH movement” and that is does reflect the dominance of lit folks. It is hard for me to see how my interests in mapping sites of social movement activism or creating “kalediscope” narratives would be enhanced by learning to code. I’m far more interested in learning to use existing technologies to do that better. I also work at a poor college with extremely limited technology access, so $ is an issue as well (my current uni provided desktop is SEVEN years old)

  9. I’ve always been worried definitely worry that attempts to push DH farther towards data analytics unsettle gender balances; this is something that comes up, in my experience, even more often when dealing with non-digital humanists who don’t want to see academic history overrun by a horde of econometricians. But I’m still committed to pushing it in that direction, because that’s where I see the interesting research questions, and that’s a big part of what I personally do. (And yet more positionally, that’s where the front lines of the old-guard humanists looks weakest to me.)

    So–I really mean this, don’t want it to sound like a criticism–what’s the positive program for someone like me whose work is, most likely, structurally bad for gender equity? You say ‘do something other’ than exhort people to code; what is it?

    It seems like what you’re suggesting might lead us to accept a split between quantitative and qualitative in DH, not overplay the macho aspects ‘digging into data,’ and head on with a somewhat gendered division of labor in a unified field that doesn’t overemphasize one type of work. That’s OK: it’s maybe even the status quo. Insofar as I want to promote work like mine, I’m doing things I would think are worse than telling people they’re inadequate for not coding. Looking down my own blogroll, for example; there are 7 men whose work involves code, 2 women whose doesn’t, 1 man whose doesn’t. (And that’s better than I feared.) Conference panels I’m on are much the same. I guess the solution is to accept a more expansive view of what my scholarly community is? And part of that is for JDH and DHQ to publish non-quantitative stuff, more? Maybe that’s enough.

    And a minor quibble: is it worth disentangling gender and race a bit more here? I’ve always thought that the degree to which DH is whiter than the humanities as a whole has to do more with digitization practices and gaps in the historical/historiographical record (it’s so easy to find massive troves of data on dead white guys, and most of them have been studied to death from other angles) than with coding culture. (Or where it is an issue, it’s tied up in gender gaps: white humanist faculty are 50-50 Male-Female, black humanists are 43-57.) I’m open to correction on that, though.

  10. Those of you who are looking for a positive program for change, allow me to suggest this (via Natalia Cecire). Please don’t lean on us to make all the change we want to see. It’s exhausting and we have day jobs, and anyway we’re all watching codeacademy tutorials after work.

    To be clear, I’m not saying we shouldn’t learn to code. I like coding. I’m saying that if you want underrepresented people involved in coding, the first step is to recognize that there’s a problem.

  11. Thanks for this post. This needs to be a central topic of conversation in a sustained way.

    I think it’s important to work on two fronts: one, we have to actively remind ourselves that there are lots of aspects of DH beyond the tech-heavy aspects. There has to be room for critique, and institution-building, and transformation of scholarly communication practices — not just for creating new tools. This matters not only (or mainly) for reasons of inclusiveness, but because we want an *intellectually* diverse field.

    But the other front is to democratize coding as much as we can. I don’t want to “exhort” people, but I’d be lying if I didn’t admit that there are huge research opportunities lying around out there, that can be scooped up by anyone with moderate coding skills and a bit of patience. I can see the opportunities lying around on the ground; they’re huge; and no amount of critique is going to make them go away. As a teacher of graduate students, it’s making me a little crazy that I can’t do more to help my students get at them.

    I am teaching coding in my graduate seminar this semester, and working as hard as I can to go slowly while getting students to the point where coding can start to feel fun and empowering. It’s challenging, though, because the culture of the humanities itself produces some of the same disempowering assumptions that you describe above in relation to gender. I don’t want to go on and on about this in a comment, so I’ll just circle back to the beginning and say that this does need to be a topic of sustained discussion in DH.

  12. At the risk of going on and on: I do worry about barriers to inclusiveness, but I’m not totally convinced that coding is the barrier we ought to worry most about.

    Coding is actually no big deal. It’s fun and everything, but it doesn’t in itself magically open up opportunities in the humanities.

    What I think *does* (almost) magically open up research opportunities is big data. I know it’s getting to where people hate to hear that phrase, and I don’t mean to say that big data is *all* we should be doing. But, in my own experience, that’s where I’ve found leverage.

    I worry a little bit that people are going to conscientiously sign up for codeacademy on the theory that coding itself is a magical key … but that this won’t really help us address the underlying structural issue, in terms of access to research opportunities. I’ll stop there for now, but try to think more about it in coming weeks.

  13. I love this post. I’ve blogged elsewhere about how many strands of geek culture exhibit unexamined male privilege which tends to be abrasive for women on the periphery of the communities… and since the privilege is unexamined, a lot of the guys don’t recognize that their unconscious behavior is part of the problem.

    Thus, I would suggest an addendum to your last sentence: So I am saying to you: If you want women and people of color in your community, if it is important to you to have a diverse discipline, you need to do something besides exhort us to code.

    …you (coder dudes) need to also do something to recognize your role in upholding a certain kind of culture that is comfortable for you but uncomfortable for others who are unlike you. Confronting privilege is never fun-happy-times, but if you’re not willing to do that much work, quit telling the less-privileged folks to get with the program and try to be just like you.

  14. Coding is a great skill to have but I feel it’s overrated and oversimplified in much of the current digital humanities discourse. Many other skills are equally important – notably a solid foundation in one’s field of specialization – to succeed as a digital humanist. And the emphasis on “coding” as a quick route to DH nirvana tends to oversimplify coding into the simplest forms of “web-development.” There is much much more to computer science and what it can offer to humanities scholarship than building nice websites, and the fact of the matter is we have to either 1>rigorously engage with those processes – i.e. learn stats, learn really involved programming, machine learning etc ourselves 2>or have a general sense of the lay of the land, the limits and possibilities of computational approaches and how humanities problems can be adapted to them without having to do it all hands-on.

    In a sense, the second one is the more reasonable and well rounded approach. One’s goal can be turning into the best collaborators for CS folks without becoming some sort of arcane hacker – and it speaks to the inherently collaborative nature of DH. I don’t expect my collaborators from CS to be up on the latest in renaissance drama, and the reverse should be true as well. A working familiarity with concepts can be gained without coding – one can read about what sort of things data-mining can do without being up to one’s ears in R (when one should be up to one’s years in whatever arcane medievalia one’s research deals with).

    I think a person who knows their field really well and has a general sense of what can be done with computational tools would make a far better collaborator and a far better digital humanist than someone who is on a shakier foundation in a specialist area but knows all about web scripting. If we need to learn anything immediately – it’s basic stats – not HTML.

    So, one way to answer the question of what positive program we might push for is to acknowledge that DH is vast and diverse and not to create a hierarchy based on one’s ability to twiddle bits. If anything, I think such a push will help those like Ben and Ted among the posters here whose work seems to be going in a code-heavy cultural informatics direction because it will distinguish between a useful and rigorous engagement with computing and a simplistic obsession about doing “cool” but trivially easy stuff with computers, and at the same time accommodate the vast majority who are capable of doing DH work without channeling them into narrow definitions of code-literacy.

  15. Last summer I was a post-doc at the Mobile Technology Research Initiative, led by Dene Grigar at Washington State University at Vancouver. There were varying levels of code proficiency in the lab where we worked for 12 weeks. A local car dealership sponsored the undergraduate MTRI fellows to design and build a mobile app, which became “Dick Hannah Customer Care.” A woman coder built the “Service” tab, and women coders collaborated significantly on MyCar, Roadside and Contact. I surmise that the WSUV Creative Media and Digital Culture program’s 50/50 gender composition reflects at least in part that women feel comfortable learning code where a woman is in charge.

    But Dene is not a programmer. She is certainly a builder, media artist and scholar. But she was learning PhoneGap literally right along side us. If we construe “building” to mean only “coding,” that’s a mistake that will limit the field and what we can do within it.

    Though I am a code novice, my experience coding (and failing, and fixing, etc.) deeply informs how and what I teach. My undergraduates don’t care if I can program. At the moment, my research doesn’t require it. Students care that I can help them collaborate to make meaningful digital objects that facilitate interactivity and/or permit interpretation. I was not a teacher who could do that until I decided to struggle with code . I agree with Ted that code is not the “magical key” that opens DH funding doors. And I don’t think that the mindset borne of working with code yields the same thing as being really good at it. But that expansion of what is possible is sufficient for me to merit some time with Code Academy.

  16. Miriam — very thought-provoking post! You’ve given voice to a very important perspective. I’m honored to be one of your two “important exceptions,” but I’d like to add a few more:
    Bethany N0wviskie
    Julia Flanders
    Bess Sadler
    Julie Meloni
    Elli Mylonas
    Dana Wheeles

    The list could (and should) go on, because I don’t think of myself as an exception, and I know that’s thanks to all the kick-ass women I have worked with in DH, including many others, including those who have never written a line of code in their life.

    I’d hate to see anything discourage women from joining DH, especially when it is the most welcoming community I have found in the Academy. And our projects are (at their best) based in collaboration — not everyone needs to learn how to code, but you do need to learn how to work together to research and communicate across disciplinary lines.

    That being said, I am all for teaching people how to code if they want to learn. And we do need more female programmers. CS used to be the most gender balanced of the Science/Engineering fields, but it has shifted radically male in the past two decades, to the dismay of many. Of course, this problem is not simply a STEM versus Humanities issue. For my first six years of grad school the women’s bathroom closest to my advisor’s office still had a urinal installed from days before the school went coed (in year 7 the History Department got a new building).

    I’m interested in learning more about @Lioness’s “Ladies Learning to Code” and am really excited about the Praxis Program at UVA. I don’t have anything like an answer to the problem, but would be very interested in brainstorming and trying something new!

  17. To Jean’s list I would add that Loretta Auvil is the CS scholar I collaborate with most closely, and she’s certainly more of a programmer than I will ever be.

    But the bottom line is that if Miriam and other early-career DH women are feeling frustrated, we have a problem.

  18. Everyone, thanks. I really need to respond in a more sustained way, but in the meantime, I wanted to give a shout-out to the women’s programming group I’ve started to get involved with, PyLadies. It’s a pretty simple idea — female Python developers teaching women who want to learn Python — but for me, it’s meant a lot. Actually, going to a PyLadies workshop last weekend was what got me thinking about why I’ve had trouble in other settings. I wouldn’t mind seeing more initiatives like this in DH.

  19. Amen, sister. Well, maybe not amen, but you know what I mean. Brava! When the STEM pipeline problem at middle school level is addressed adequately, then I will feel much more friendly toward the notion that writing code is a pre-requisite for participating in digital humanities.

  20. The coders (hate that name) do not have a glamorous job, even though it pays reasonably well. We are the plumbers, the housekeepers. We unclog the toilets. We make the data flow through the pipes. My impression is that a lot of academics (male and female) sense this and choose not to become *too* adept at programming, because if you’re the person who can fix it, you’re going to be the person who fixes it. It’s better to be the person who tells other people to fix it (and writes papers and grant proposals, goes to conferences, etc.).

    So there’s more than just gender and race axes here, there’s status/class issues too. Why would a smart, capable woman choose to acquire a skill that’s a) hard to learn, b) where you aren’t sure you’re welcome, and c) will bring you less recognition than your colleagues get? Especially when recognition is the primary currency in academia.

    I really, really hate the gender imbalance in software engineering, and I hope it improves, not least because I’d like my daughters to feel welcome there (should they choose to go that route). But it’s a complex tangle of problems. Programming *is* a boys club, with all the irritating features of a boys club, because it’s mostly boys. If we can adjust the balance, it will cease to be such a boys club quite quickly. The trick is getting a lot of women to start doing it. And there’s still the status question.

  21. This is a big, complicated problem, and for now I just want to address one little corner of it. Miriam is quite right about the gender/class/race hurdles and I applaud her for working to push through them herself. I am surely no coder by professional standards, but I have a background in it and can muddle my way through what’s needed, if slowly and painfully at times.

    In the conversations I’ve been part of, the question of code and whether it’s necessary most frequently comes up as a question of curriculum–should the teaching of code be part of a digital humanities (usually intro) course? And I’ve been adamant that it should. Of course, by quirk of career I haven’t taught any digital humanities courses since this all became “digital humanities,” but I was teaching first year students to code basic HTML in 1995 so that they could, pre-Dreamweaver or WordPress, post their papers on their Web sites. And I was teaching them to MOO, or program in MOO if such captured their interest, as part of some early virtual classroom experiments. I felt, and feel, that there are two important reasons to do this, teach the code in the context of our humanities + digital courses. They are: 1) The context. I want my students to think about how computing extends or could extend their explorations of the humanities content we’re working with. My specific code examples, say an if/then loop or what is an array, may be similar to what they get in their first weeks of computer science, but my problems and questions are not (econ number crunching vs. a poetry slam machine, let’s say); and 2) The possibilities. Having worked with people on digital stuff in all kinds of settings, I’m firmly convinced that those who know a bit about how computers work are better able to imagine what those computers might be able to do for them. Even if they just come to understand that they could possibly control, with appropriate knowledge, what computers are doing… otherwise, we’ve given away the keys to the Web site/database/castle–in the form of syntax or passwords or UNIX permissions–because we didn’t know they were ours to keep.

    The other point is that “code,” for me in this discussion, is a shorthand or codeword for knowing how to apply algorithms. The basic thing I want people to grasp is not the syntax of this or that language (although knowing that they each have their syntax and that it can be crucial in getting code to function is important), I want them to understand the basic process of manipulating code/data/text and to have a sense of the basic toolkit of available options for accomplishing it. So, again, the basic if/when loop. It gives me a way to compare one word to all the other words, or to add stuff together until it’s all been added, or to do those or something else for just awhile until I give it a reason to stop. Or print or say or display or whatever it is in your language of choice: it gives me a way to have words appear on the screen, possibly as part of a timed script that interacts with a user.

    All of which is to say, I don’t think the ultimate answer is to give up on “code.” A better answer is for the digital humanities community (however we’re defining that) to find ways to include rather than exclude, both in terms of what we expect or require of each other and also in terms of how we share our knowledge with our peers or teach it to our students.

  22. It’s been a while since I’ve thought through these differences. But if, as you claim, there is an exclusionary aspect to coding, it’s also worth acknowledging how much coding is like writing. The following from “Code and Composition:”

    the differences between spoken/written languages and coding languages serve to cleave the activity of composition and the activity of coding into separate cultures in ways that resonate with C.P. Snow’s complaints of 40 years ago. But if we can look past these differences, many more commonalities exist than some people would initially presume. Coding doesn’t expand one’s emotional universe as readily as writing. But this doesn’t mean that the activities are alien—in fact they should be familiar to one another. Programmers (like writers) are attempting to create order and meaning out of disorder. Programmers (like writers) also have to negotiate transitions between social worlds and more quiescent worlds where the code is actually generated. And programmers (like writers) also scribe as a way of channeling power. Writers and programmers might not have occasion to intersect much as they make their separate ways across campus. But coders and composers share a common set of tools, techniques and challenges in their quest to fashion order out of disorder.

  23. Thanks for this post, Miriam.

    If DH is about making, then programming may be one of several tools in a digital humanist’s toolkit. Programming is not, in my opinion, an essential skill for all humanists. Don’t get me wrong — I think it’s useful, but I’m not going to make a “DH is ___” claim here. Making critical decisions about what data to use, and understanding the basics of what a tool is doing under the hood? Those are increasingly important, and the degree to which that applied knowledge is connected to programming may lie at the heart of this “should DHers program” question.

    Regardless of my own perspective on the value of coding, pointing out the perceived importance of programming knowledge as “the word on the street” is valid. The skills that I pointed to are not easily quantified, and it’s easier to refer to a GitHub repository than it is to determine the appropriateness or validity of using digital technology in one’s own research. In my opinion, this latter engagement remains largely invisible and undervalued in DH.

    I think we need to push the topic of diversity in DH further. As Jean pointed out, CS departments are radically male, which is frustrating because even within CS there have been active attempts to overcome structural problems that may cause this. I worry that as DH becomes institutionalized, this problem will become hardened in negative ways. I hope to see this conversation continue.

  24. I want to go on record by saying that I code. I have been coding since 1994 when I made my first website using command line programming to do it. I published the first research webpages in Dallas during that year. When HTML emerged, I taught myself how to code with it by borrowing other people’s code and tweaking it. I did not know how to make apps, so, yes, I joined in the courses last summer to learn. So, Kathi spoke incorrectly when she reported that I do not know how to code. I now managed over 9 websites. Some, like my personal page, are coded by hand; some have been made with WP, and others with iWeb. My university program site is hand-coded, and I hand-coded all of the content for the MLA12 Electronic Literature website. But even with CMSs I find it necessary to go back to the code and tweak the pages by hand. That is where the power is–in the code.

  25. I apologize to Dene for suggesting that she does not code. I should have worded my statement more carefully to draw a distinction between coding to build digital artifacts (what Dene describes of her own work) and being a programmer, which I take to mean writing algorithms that, in the context of DH, allow processing of “big data”: the sort of work Ted Underwood mentions in his post. It was a sloppy mistake not to draw that distinction.

    Dene is absolutely right that “that’s where the power is — in the code.” She earned her cred in the command line, and has built an influential and generative body of work enabled by her code expertise.

    One wonders if the gender composition of programmers and coders in DH were to be come more balanced, would programming and coding lose value? Is the emphasis on programming’s materiality insurance against DH becoming a pink collar ghetto? [See Elizabeth Abel’s 1993 discussion of English departments as “pink collar ghettos” in which tenured white female academics’ seek to recuperate a lost subject position of disenfranchisment by writing about texts by persons of color. “Black Writing, White Reading: Race and the Politics of Feminist Interpretation” [Critical Inquiry Spring 1993.]

    One “Critical Difference” that inheres now that didn’t when Abel wrote is that women DHers — like Dene, Tara McPherson, Elizabeth Losh, Bethany Nowviskie, Lori Emerson and others — run labs and programs. These labs can create a critical mass of practitioners for whom a defamiliarized vantage on computation and the “imbrications” (McPherson) of race, class, sex and gender is elemental to understanding what we make when we make things with code.

  26. To avoid misapprehension: I suspect considerably more coding skill is required to build a website from the command line than is required to do anything w/ “big data.”

    There’s nothing technically challenging about “big data” on the scale that most humanists are concerned with (which isn’t all that big). You do have to deal with the fact that it can’t all be memory-resident at the same time; so you store it in MySQL or something. But that’s basically it for technical obstacles. There may be obstacles of other kinds that we have to think through.

  27. Trying to FORCE women, men, dogs or cats into coding is an exercise in futility. I am in marketing, and if someone ever tried to force me to do that my head would explode. Trying to impose a foreign discipline on the unwilling or uninterested (assuming all are able) doesn’t make sense and will produce shoddy product. If you know a coder, ask them how excited they are about taking over someone’s product riddled with shoddy code. Waste of time, money and mission. Seems like social/industrial engineering gone awry.

    You want -people- to code? Give them tangible incentives beyond increasing skills or having something for the resume. Give them the opportunity and the chance for promotion, additional responsibility, autonomy, or some other sort of material benefit. We humans are pretty simple.

    And a side nerd note: Lots of HTML mentions here. I think it’s worth noting that many developers do not consider HTML to be “code.” It’s actually considered “markup.” I know and have coded HTML5/CSS3 by hand, and I by NO stretch of the imagination would ever label myself a “coder,” “programmer,” or developer. Kind of nitpicky and definitely disputable, but putting it out there.

    And how’d you know I learned BASIC when I was 10?

  28. There’s a great conversation here. Many or all of these issues have come up in more formal studies of gender issues in software development (like the ones done under the auspices of the ACM), but Miriam’s post is really clear and succinct, and I think this is an important issue for DH as a field to grapple with.

    One problem is the rather amorphous goal of “coding”. Some of the responses touch on this. Coding isn’t a skill; it’s a wide range of skills. Saying you can code is like saying you can build a structure. It’s good to be able to build a lean-to out of found materials; it’s good to be able to design a skyscraper; it’s good to be able to install a watertight composite-shingle roof. But those are different skill sets, and we don’t expect someone who’s good at one to be good at the others. So it is with writing software. I’m a professional software developer, a computer scientist, a digital rhetorician, and on occasion a DH scholar – and there is usually very little overlap in the kinds of programming work I do in those four roles.

    Asking DH scholars to “code” drops them into the middle of the sea. Programming is difficult enough as it is, with its arbitrary rules and many levels of abstraction and random reinforcement and other psychological traps. Asking someone to choose from a huge array of possible competencies to acquire exacerbates that, and the more social pressure that person is under (say, because she’s viewed by the programming community as the champion of an underrepresented group, as Miriam mentioned), the worse the experience will be.

    And while coding is a useful skill, it shouldn’t be mandatory for DH work. If it is, we haven’t done a very good job creating tools for DH. (And we haven’t, really. We haven’t done a very good job creating software in almost any area.) There are lots of useful skills; we shouldn’t become obsessed with one subset.

  29. This is similar to what a few others have said here but I think this is a great moment to really stress that “DH is about making things” and “DH is about writing code” are very different ideas about what the field is and should be.

    I am not, nor will I ever be, a programmer. With that said, I love working on software projects; I love UI/UX work, I like doing outreach, I like doing software documentation, I like teaching workshops on software and I like acting as an ombudsman and advocate for users in the development process. In my mind, all of these activities are doing DH, all of them could and should develop their own DH scholarship and writing and, for the most part, all of these things are left on the table when DH means “learn2code.”

    This is not to say that there isn’t considerable value in learning to code just to say that it is one of a series’s of competences that one should want on the DH team.

    Lastly, I think the continued success of DH is going to be largely about some people making tools and many more people using those tools to do interesting work. For example, you can use Voyant without any coding chops at all. Now, if a tool like Voyant is going to actually be useful it needs to be used and figuring out what to do with it, doing that, and thinking about how these tools could work better in the future is invaluable DH work. Frankly I think part of this is about getting more user friendly tools out there and building user communities around those tools.

  30. This is a fantastic conversation, and I’m thrilled to see the ways that it has developed from Miriam’s initial (amazing) post.

    I want to pick up on something Trevor mentions: that “making things” is a value that includes but is far from limited to “coding,” and that a lot gets left behind when we obsess about code. There was recently a lengthy discussion on another site that focused on what was seen to be the silly tendency of DH to think/write/publish about itself, which it was urged to stop, so that it could get on with the more important business of doing things. Lumped into the category of not “doing” anything was a whole swath of the kinds of work that Trevor described, including building communities, encouraging folks to work in new ways, teaching other folks how to engage with that new work, and so forth.

    All of these tasks are, as Trevor rightly points out, crucial aspects of making things. I want to highlight this not just because those are the tasks in which I have personally been engaged for the last six-plus years, but because there is often a decidedly gendered division of labor in project teams; the more that the locus of value in making things is restricted to coding, the more likely it is that women will be relegated to the role of support staff, with their crucial contributions seen as window-dressing.

  31. Great discussion, all around. This is a little off-topic, but re the recent comments by Trevor and Michael: I agree with the spirit of both comments. But … so as not to mislead early-career scholars … I think it’s important to admit why we don’t yet have better tools. It’s not (just) because UI design is hard. It’s more crucially because we don’t know yet what kind of tools we should be building.

    At least in the text mining domain, I suspect we really haven’t figured out yet what text mining is for. We may think we know. But the field is still very fluid, and many of the tools we have borrowed from CS may need to be revised or extended for humanistic purposes. So while I have the greatest admiration for toolkits like Voyeur and MALLET, I wouldn’t advise an early-career scholar who wanted to practice text mining to limit themselves (over the long term) to the forms of analysis possible within those toolkits.

    Or to put that more bluntly: knowing how to code is important because we really don’t know what we’re doing yet, and fiddling around with code may be the only way to find out. 🙂

  32. Ted your example is a great one to pick apart some of the key bits here. Yes, if your goal is to do cutting edge text-mining then you are not going to be using Voyeur and MALLET. With that said, doing creative things with Voyeur and MALLET and coming up with novel ways to get them to do interesting things are doing valuable DH work. Similarly, if someone wants to do user testing on these tools with different groups of scholars they can produce valuable contributions to DH. If someone wanted to mock up different kinds of interfaces to these tools they would be doing valuable DH work. Refining how to teach workshops on these tools would be valuable DH work. There will always be cool things that easy to use common tools can’t do, but that shouldn’t stop us from foraging ahead and working on other sorts of DH scholarship.

    This is to say that doing the humanities digitally, that is using technology instrumentally to answer the kinds of questions we have always asked is important work. With that said, it is just one part of an ecosystem that is emerging here. You are right to note that what exactly a text mining tool package should do is itself an ill defined problem, but the same is true for just about every other exciting part of this space. I would also mention that things like Google-n gram and or Mark Davies Corpora provide amazing ouppertunities for us to try out all kinds of interesting historical text work on some big bodies of data and only require one to spend an afternoon reading documentation and playing around with them to end up thinking up interesting, novel, and creative ways to use them to answer humanities questions.

  33. I have taught programming and software engineering to graduate and undergraduate students in the humanities for about ten years. As a teacher, I am very passionately devoted to the subject and to the class, and coding is at the center of most of my research. In most respects, my classes follow the basic curriculum of first- and second-year computer science, but with a focus on the sort of data that humanists care about (mainly, though not exclusively, text data). It also adds lengthy discussion of the historical and social elements of computing. So discussions of control structures and variable assignment are interwoven with discussions of online identity, transhumanism, gaming, social media, and debates about artificial intelligence.

    But here’s the thing: With a few exceptions, the students in that class have been majority women for nearly the entire time I’ve taught it.

    And what I see in those classes — year after year — are unbelievably talented women who discover, to their astonishment, that they’re not only ferociously good at coding, but that they really like it. I won’t say that all the women who take my class feel this way (not all the men do either), but an incredible number of them do, and many of them are moved to go on. So after taking my class, they often head over to the CS department with the thought of continuing on in the subject.

    And there it ends. Not one, to my knowledge, has ever gone on to major in CS, or even to minor in it. Many can’t make it through their first class.

    Now, let me be clear: We are not talking about students who lack the aptitude for the subject. My class (which is at 400-level) is more intensive than the introductory sequence in CS, and the women who try this are the best people coming out of it. These are women who have fallen in love with the subject and who are highly motivated to continue. I don’t know if there’s such a thing as a “programmers brain,” but if there is, these women have it. They show every sign of going on to become highly accomplished engineers. And then they stop.

    Why? The stories they tell are so disheartening, they make me want to hang it up completely. I hesitate to summarize the experience of a lot of individuals, but I’ll risk it here for the sake of illustration. They all blame the “culture” of computing. Not all of them (may are first- and second-year students) can find that word, but that’s what it amounts to. They insist it’s not the professor, who more often than not is going out of their way to be welcoming and supportive (though that, as we all know, can be threatening in its own way). But they hate being the only woman in the room. And they hate being subtly ogled (“Look! A chick who’s into computing!”). And they hate the do-or-die, get-it-or-you-don’t tone to the whole thing. And they hate the snarky, sexually-tinged jokes. And they hate the thoroughgoing emphasis on being some kind of cerebral black belt, to the exclusion of any other kind of way to be.

    I witnessed this first hand the other day. We’re holding a development contest at my institution, and I encouraged my students to enter. One woman, a really talented hacker (who had never programmed before my class), was particularly interested in joining, and they were holding a meet-and-greet so that they could explain the terms of the contest and get people into teams. I told her I’d come along.

    It was like a locker room. I counted three women in a group of at least fifty men, but that wasn’t even the worst of it. Porn joke? Check. Sports and warfare metaphors? Abounding. Do-or-die, you-win-or-you-suck vibe? Very much in evidence. And there’s my wonderful, brilliant sophomore taking it all in. I was so pissed off, I wanted to cry.

    And I understood, in a way that I hadn’t before, why my students had so much trouble putting their finger on what’s wrong with all of this. Because by some measure, nothing terrible was happening. Everyone was excited, the people leading the project were enthusiastic and supportive, the whole thing was very student-focused, and there was a general sense of esprit de corps emerging. Everyone was eager to make this a fun thing. I am certain that if you asked any of the men in the room — whether faculty, students, or staff — what they think of women participating in this contest, you’d get exactly the answer you hoped for. It would be full of every kind of piety you can imagine. But if you’re a women, everything about this says that you’ve wandered into the wrong restroom.

    Walking out with my student, I tried to say what I was thinking, but since what I was thinking was, “Please, please look in your heart and tell all these guys to go fuck themselves,” I found it really hard to verbalize it in an appropriate manner. I mumbled something about there being a frat-party atmosphere in there; she mumbled something about the high testosterone level. I was just heartbroken. Because even if could manage to give her the kind of pep talk necessary to overcome this (which I can’t), I’d be hard pressed to figure out how to change that culture. And it’s my culture, even if I’m as appalled by the porn jokes and bullying tone of the whole thing as I suspect most women would be. There’s just not a lot that a middle-aged guy can do for nineteen-year-old women hoping to overcome these obstacles. I can strive not be that asshole, and I can be encouraging and supportive (without the sickening “Oh look! A girl!” nonsense), and I can set these women on the road to becoming highly skilled developers, but at the end of the day, that just doesn’t seem like much.

    I am a well-known booster of the idea of “building” in DH, but I have always hastened to add that building does not necessarily mean coding. I don’t expect everyone to be good at coding. I don’t expect everyone to like it, and it’s hard to get good at something you just don’t like. Like most people, the list of things I am bad at and don’t like is terrifyingly long, and occasionally embarrassing. But what I can’t tolerate at all is the idea of someone — whether a student or a colleague — who both likes it and is good at it, but is turned away from it by the “culture.”

    Fuck that culture. I’m a teacher. I care about students who want to learn, learning. I’m not so naive as to think that we can reform that culture from without, but honestly, if we just re-duplicate that culture in DH, then we have failed. And we might as well go back to whatever we were doing before.

    Sorry for the rant, Miriam. This is a great post (even if it’s a hard one to hear yet again), and the discussion thread is giving me some kind of hope, even if it remains more-then-slightly terrifying to think that we might be doing it all over again.

  34. This is a fascinating thread. I’d like to recommend Wendy Chun’s book *Programmed Visions*, which I’m reading at the moment. It considers the myth of code as power, and particularly relevantly describes the gendered world of early coding. It’s one of the best academic books I’ve read in ages.

  35. Hi there.

    Found this post via twitter and I have to say I’m quite astonished. I learnt (ZX) basic at 11. I did computer science in my first and third years at university, back in the late 1980s/early 1990s. And I don’t recognise this as a problem, either at university or once I went into programming in 1998. And it’s not that I wouldn’t recognise it – I am female. In many cases I’ve been the only female on the team, and I’ve interviewed in a number of places where I’d have been the first woman into the dept. At one place this was viewed as such a positive that they entered a bidding war with my then employer.

    Could this be because I’m in the UK? Do we not have such a negative atmosphere? Is it because companies here have stricter rules about the type of behaviour that you describe? Or have I just been incredibly lucky?

    I have experienced discrimination. I have been told that I was being aggressive, when I’m fairly sure a man would have been praised as assertive and ambitious. But that was a company issue, nothing to do with an IT environment. And I did attend one interview where the two interviewers tried to outdo themselves with unpleasant language and extreme male attitudes, but again, I put that down to poor management rather than it being related to IT. Am I fooling myself?

  36. My mother was a programmer until she retired, so in our household programming was seen as a gender neutral thing. I enrolled in computer studies at the end of high school, excited at the opportunity to get involved with cutting edge technology. However, I found it difficult. The boys had studied programming for 2 years at their boys high school but our girls high school had not offered anything. I entered the lab to start my first programming assignment and was confronted with a room of 16 and 17 year old boys who had finished it and were embellishing it for fun. They were intent on their work and not interested in giving me a go at one of the few terminals in the room. I pushed in and then had to endure them looking at my feeble attempts and giving ‘advice’. I gritted my teeth and stuck at it and by the end of the year I was enjoying programming. I continued to enjoy programming at university as part of my accounting degree, still in the minority, but the boys had grown up and we became friends who supported each other.

    Fast forward about 20 years and back at university doing a history major this time. I noticed that the young students around me assumed that I had no competency with computers because I was female and middle-aged. I realised that in a few years time I might want a job from them and they just wouldn’t believe me when I told them that using technology was a core part of my entire professional life. I had to have a ‘qualification’ on my CV. I enrolled in a computer science unit and was once more in a male-dominated computer lab.

    The lecturer made a big fuss of having female students in the room and gave me extra attention in the lab which was unneccessary and made me feel very concious that I was one of two representatives of the sisterhood in the room, rather than a student like everyone else. BUT the male students included me and we worked together well. I didn’t experience any male competitiveness or banter. I had the added burden of being almost old enough to be their mother but that didn’t seem to be an issue either. I have since attended a THAT Camp and followed the digital humanities crowd online. Never have I felt an outsider because of my gender or my age.

    Yet it is disappointing to see that between the time I first went to university and my most recent student experience there was no improvement in the proportion of females doing computer science. My daughter was the only girl who did electronics in her year at school and my husband’s software firm had one female developer in the entire development team! This demonstrates how little progress we have made on the gender issue in programming (have we gone backwards?). Something has gone wrong.

    At school when students have not matured male banter and competitiveness is a problem and a turn-off. It is at this age when socially it is established that technology is a male thing. I have never experienced a problem with male banter and competitiveness excluding me in a programming environment since I left school. Is this an American phenomenon as suggested by liveotherwise? I am reluctant to say so as in general life there is plenty of this type of male behaviour here. Maybe my personal experience in university computing labs is not representative? Rather, I think that the mores established at school, about what men and women do, become set in concrete as people grow older. Male students find a male computer lab normal, the thought of doing computer science units would not even cross the mind of most female students (programming is not offered by humanities departments).

    And so the pattern established elsewhere can easily filter into digital humanities unless we take active steps to counter this trend. I reckon initiatives such as the PyLadies group is a great idea.

  37. Fascinating and important conversation going on. Though it is not so easy to measure the influence or impact of cultural norms on our everyday life choices, I think Stephen Ramsay’s post highlights the numerous, and often subtle, ways in which women are not made to feel encouraged, entitled, or enthused to code. Paying attention to the DH spaces we are creating, producing, and sustaining would probably be a helpful place to start – but also injecting some of the enthusiasm and mentorship that helped the community grow might also be good.

    I can only speak for myself, but as a self-taught “technologist” also a product of the 90’s DIY ethos, I felt initially both inspired and capable working in the tech industry. I was hired and trained to be an Oracle database programmer and worked at a great place for eight years, leaving to become a tech consultant, and then a Digital Asset manager. Through the course of 10 years or so, a series of things happened which I thought I would mention: a few too many run-ins with higher ups who would rather me take notes at meetings than lead them; having to work with, and face sexual harassment from, male network administrators; moving to a geographical region where I applied for countless jobs without receiving a single call back until I changed the name on my resume to be gender-neutral, subsequently being offered a job as an Oracle DBA. At some point, I decided it would be helpful for me to pursue a master’s degree in Information Science, because I had come to feel like a self-taught hacker. I’m sure this is not unlike the imposter syndrome so many female academics describe.

    The digital humanities community initially appealed to me because I saw opportunities to “make and do things” in innovative ways. More active mentoring and commitment to, and value for, diversity in DH are important and good. But I think we could also try to envision new ways in which we as a community can address these (very old) problems. After all, how can we truly create “new ways of knowing” if we just end up hatching the same old structures again?

  38. This is a wonderful and much-needed discussion. I had the same reaction after reading the original post but wasn’t sure quite how to respond. Thanks for this.

    I’d like to second Michael Wojcik’s comment above that coding is a wide variety of skills. I think it’s problematic when we disavow the identity of “programmer” simply because we don’t fit the definition of “programmer” in CS–i.e., we code in html or other “mark-up” languages or can’t build compilers. It’s important that humanists own their own definition of what it means to code. Coding is too important to leave it in the hands of CS. So by my definition, I code. I will never code according to CS, but I’m an English professor, so I’m fine with that.

    I’d also like to underscore the story that Stephen Ramsay tells. Yes, it’s a locker room. It’s all in good fun, but there it is. It doesn’t affect me much now because I know what I’m about, I have a career, and I’m in my mid-30s. The point that Miriam makes above about learning BASIC at 10 is critical: at several earlier points in my life, I turned away from coding because of the culture. But I didn’t know that’s what it was. In my teens and early twenties, I attributed my turning away to disinterest in coding itself.

    Finally, I’d like to recommend the work of Nathan Ensmenger, who traces this masculinization of programming. Programmers were often women in the 1960s (the men were engineers), and women were better represented as CS majors in the 1980s than now. No, we’re not there yet.

  39. I’m one of those “old-timers” who learned to code BASIC at age 14 on a thermal paper machine with a cradle modem. So returning to coding after decades because of DH does have a nostalgic quality that has a distinctly gendered aspect. But there certainly seems to be a generation shift afoot as well. Working on a project involving data viz. and R coding I found that my bibliography was almost entirely male, but my research assistants (at Emory and GaTech) were overwhelmingly female.

    Details aside, thanks for the “consciousness raising.”

  40. This subject comes up a lot in my household: with two daughters born of two software engineers and with my wife’s second major in women and gender studies, obviously we’re concerned about women in tech. I could do a whole lot of mansplaining here–hey, some of my best friends are female software engineers!–but I’m going to limit myself to one point: university technology projects are tremendous agents of economic mobility and I want more women to participate in them.

    My wife and I each followed the “traditional” CS path — personal computers in the home when we were in elementary school, computer science majors in college, followed by professional software engineering careers in IBM and venture capital backed start-ups. However, a lot of our training–the resume line-items that made us a bit more attractive to employers and the work habits that made us a little more successful in our first jobs–came not from our coursework but from campus tech jobs: phone support in the campus help-desk, student UNIX system administration, running the mainframe that powered the library card catalog system, or website development for research groups or art history courses. But we both had computer science degrees — what about everyone else?

    I’d argue that the economic effects of on-campus tech roles are even greater for people who didn’t come through that narrow CS-grad pipeline. Many of my coworkers in technical roles came from the local university IT system with no CS background. One of them was a history PhD who’d tried to cobble together a career out of adjunct positions while paying the bills doing data analysis for the university. That experience got him a position doing data integration at my employer five years ago where he has been promoted four times and is now a second-line manager over our technical services teams. People who are not me may draw conclusions about his prospects in academia — I’m just glad that his data analysis projects gave him another option.

    Most readers of this comment thread know a lot more than I do about the humanities job market. What I do know is that tech companies here in Austin are desperate for people. (My local Ruby on Rails user group now is able to rent out the top floor of a downtown bar, with catered food and an open bar funded by sponsors who are trying to recruit people.) However, they’re looking for people with a bit of experience who are smart and can get things done. DH projects can get people past that bar even if they don’t have a CS degree, come from a childhood of hacking BASIC, or have the privilege that gets them through the gauntlet of under-30 programmer culture.

    I could go on for pages here — about how we define success, about DH as a “second pipeline” (see Sara Brumfield’s 2009 Lone Star Ruby Conference lightening talk), about how to find a healthy code culture, about critical masses and 4% female conference audiences, or even about programmers not having the rhetorical skill to express their own experience [mansplain alert!]. But it’s probably wisest to quit by saying: please don’t discourage women from learning to code.

Leave a Reply

Your email address will not be published. Required fields are marked *