This an Ask a Geek Feminist question for our commenters. I have some comments on this one at the bottom, but not a real answer.
I’ve seen various mention of trying to increase the respect given for non-coding activities, such as documentation and testing, which seem have a better gender ratio than coding, as a way of increasing acknowledged female involvement in FLOSS. But, while we definitely should give more recognition to non-coding involvement, it seems to me that allocation of respect / recognition simply does naturally concentrate on that which has the longest and steepest learning curves (just as I guess that in running there’s a hierarchy of jogger – runner – marathon runner – hypermarathon runner), and that this route will risk perpetuating a division into “womens’ work” and “men’s work”, with the traditional difference in public valuation. Is this a risk? Is it happening? And if so, what can we do about it?
And likewise, I get a similar impression about scripting vs compiled languages — that, statistically, women (more so than men) tend to prefer languages like python, rather than the languages that they’re implemented in (typically C). Is this a real divide? And does it have risks of getting more female involvement in FLOSS but in a way that some [male geeks] will dismiss as “not the real thing”?
Something I think is worth considering about this question is whether or not the hierarchy the questioner gives is objective. I’d argue that it largely isn’t. The learning curve for coding can be long and steep, yes. But consider documentation, for example. Writing well is a really difficult skill. It’s sometimes not as obvious that you’re acquiring it, precisely because it’s such a very long process and it involves doing a lot of reading and practising other forms of communication as well. A baseline level of skill in writing is also more common than a baseline level of skill in coding, but a high level of skill is no easier — I’d actually guess much harder at the very extremes — to achieve.So we need to be very wary of accepting this hierarchy at face value, both because it buys into the existing undervaluation of certain skills and because it risks continuing a nasty pattern: “if women can do it, it must be easier than we thought, let’s look for something currently mostly done by men and value that instead.”
That said, coding is fun and useful. (Well, for me. But that’s enough!) So is nuclear physics, pure mathematics, electrical engineering, hard SF and many other “male” halves of the gender binary fractal. So we don’t want to cede those to men.
For more of my own thoughts on this, see “Girl stuff” in Free Software, a post from last year from the point-of-view of deciding what to work on as a woman. What do you think? Where’s the balance between creating and properly valuing roles more suited for women’s existing socialisation and more women entering male-dominated and currently highly valued roles?
Geek Feminism Blog 
This is, I’m afraid, a bit of a rant and definitely off-topic.
Speaking as a tech writer, what people don’t get is that there are (at least) two different skills involved. Writing, sure, but also researching, and researching is just as hard. In fact, most experienced writers I know agree that once you’ve done the research, the writing is a piece of cake. Why? Because often the author of the code doesn’t think like a user; he or she thinks like its designer. The code itself can be an unreliable witness, because many people out there’s code is strictly write-only, and in any case won’t answer questions like “How many interactions per second is this built to handle?”
Over the years, most of the engineers I’ve worked with have been surprised and delighted because I actually wound up helping them understand their code in ways they never had. I asked them about boundary cases they hadn’t considered, I asked them “But which way do you *want* people to use this?”, and I asked “There are two subtly different methods on this class, what’s the use case for each of them?”
To do a good job of tech writing, you have to have top-notch research skills, and to do that you have to have a good understanding of what the software is supposed to be doing. A good tech writer, she said modestly, is a big ol’ geek with an insatiable curiosity, as well as those writing skills you see mentioned in the title.
As to how we get respect…. well, the best way I know is to edit something — anything — for a fellow engineer. Once s/he sees the difference between the first draft and what you can do to help get her/his thoughts out there in coherent form, you have a slave for life.
I suppose what the world needs is an Open Source Editing project.
Related — I think this is a good example of the gender binary fractal posted here in January — awarding geek cred along lines like scripting vs. compiled languages sounds just like dark blue vs turqoise to me: a desperate attempt to redefine geeky so as to *stay* axclusive. (much like the “hardcore gamer” definition attempts!)
Jogger / runner / marathon runner /
hypermarathon runnerultra runner (longer-than-marathon events are called ultramarathons) aren’t accorded ever-increasing amounts of respect. Those who run fast 5K / 10K / half-marathons generally get far more respect than the average marathoner (after all, most runners can finish a marathon, if they decide that they want to put in the time to train for it, but most runners can’t run fast). Ultrarunners are generally considered to be deviant and of questionable judgement, and are thus removed from the respect scale altogether.I don’t think there is a risk of anything in free software being ghettoized as “women’s work” (ie, to the extent that men would be wary of being associated with it) because the number of extra women needed to create that perception is off the charts. That’s what it seems to me.
You know I would have thought that too, but I have been going to my local Python user group for over a year now and I haven’t yet seen a single other woman at a meeting.
The Linux.conf.au conference has hundreds of the most hardcore geeks I’ve ever met. The most popular self-nominated Linux distro from the attendees? Ubuntu.
Arduinos were originally designed for artists and non-programmers, but have been taken up enthusiastically by all manner of male geeks, who seem to value the potential for creativity over the potential stigma.
It’s not that “scripted” (sic – did the OP mean “interpreted”?)/dynamic languages are “not the real thing”. It’s that whatever women do, some men will dismiss. The “I’m more hardcore/l33t than you” ego game is not worth playing IMO.
Besides, what is the alternative? Force newbies to learn C? Then the goalposts will move – you mean you don’t write assembly?
OP here (who mostly uses interpreted Lisp for pleasure and compiled C to earn a living) — yes, “interpreted languages” is pretty close, but it’s actually a bundle of language characteristics that often go together, variously described as “Ousterhout’s dichotomy” (or “Ousterhout’s fallacy”). Could it be that women with lowered self-efficacy find languages seen as “easier” more attractive, at least as newcomers? Which may be fine if seen as a “gateway” language as Amberalla pins it down, but it’s a pity to stop at that. But I’m sure you’re right about some men dismissing whatever women do.
Perhaps the alternative (also see Sophie’s post) is to teach people (at least, those formally learning CS) several languages, with the idea of enabling them to choose an appropriate language for a given problem? Teaching several languages would be good for other reasons too, including more understanding of the breadth of what can be done; there seem to be all too many CS graduates who are only taught Java and don’t expect to have to learn anything else after graduation. And one I’ve never heard of: teaching documentation as part of a CS course? But these only apply to those learning CS at Uni etc.
Hi OP, nice that you stopped by to see your replies. :)
I definitely agree that learning multiple programming languages will improve any programmer’s skills. But it seems you are troubled by the idea of “stopping” at scripting languages. “It’s a pity to stop at that” – why? It’s not like there is any shortage of wonderful, amazing things one can do with scripting languages. I think Ousterhout’s claim might be rubbish. Sure those languages are great for writing one-off scripts – but they are also great for writing lots of complex systems. Maybe not a kernel or an OS (although I think much of GNOME is written in Python…), but why should we hold those things up as the holy grail?
I don’t think that focusing on non-coding activities is an effective strategy for recognizing and promoting women’s involvement in software, for a couple of reasons.
The current imbalance has its roots in inequality, and in order to change the status quo, that inequality needs to be recognized and changed. If the conversation is about functional roles, rather than gender roles, it misses the fundamental issue.
While I agree with pfctdayelise that there is little chance of ghettoization, there is the risk of implying that women are somehow better suited for non-coding work, which creates barriers for women who do code, and discourages others from choosing such a role.
I’ve seen no data which support the hypothesis that higher level (“scripting”) languages tend to have a better gender balance. Is this actually the case? What other factors might correlate with this one, which might be more fundamental than gender (length of experience, subfield of programming, such as web app, native app, operating systems, etc.)?
All of that said, I do think it’s important that the software community integrate more of the non-coding activities which contribute to the creation and success of software. This is a good idea, but not (in my opinion) on feminist grounds.
Exactly. The point of feminism is choice, and valuing non-coding work, or scripting languages, has nothing to do with feminism, unless you believe that women are innately more suited to non-coding and scripting languages, while men can do anything and everything.
Simone de Beauvoir complained that putting women on a pedestal for doing housework is actually used by men to reinforce the patriarchy:
I prefer coding, but at work, my boss tried to steer me into doing more documentation and secretarial work, because he assumed that I preferred it to coding, probably because I’m female.
Also, many women than men may be using Python over C, because Python is a new language, while C is an old language. In the past, the gender disparity in computer access was more pronounced.
As a woman who prefers Perl, I find the scripted/compiled question interesting.
In the end, though, if the language is Turing-complete, it doesn’t matter – all languages can *do* exactly the same things as any other language, which – to simplify things rather massively – means that the difference is how fast the resulting program runs. Modern interpreters can all do just-in-time dynamic compilation, making it almost as fast as native code.
Obviously there’s more to it than that; each language also has its own specialist areas, so it’s easier to code some things in Perl than it is in C, for example. Some languages have a highly accessible library/module repository, making it easy to benefit from others’ code. (Perl has CPAN, for example, and PHP has PEAR.)
What I’m trying to say is that there *is* no ‘perfect’ language; most are comparable to each other in their output anyway.
“The current imbalance has its roots in inequality, and in order to change the status quo, that inequality needs to be recognized and changed. If the conversation is about functional roles, rather than gender roles, it misses the fundamental issue.”
I don’t think it’s appropriate to respond to “women’s work is stigmatized” with “women ought to do men’s work.” (Where by “women’s work” I mean “work that is dominated by women” rather that “work that is right for women”, and vice versa.) Some women genuinely enjoy the non-coding and, for that matter, non-kernel-oriented coding jobs, just as some women love teaching, nursing, and so on.
I’m not down with “Promote lower-status geek jobs to get women into the community”; however, that’s a different statement than “Celebrate the lower-status geek jobs that many women already do.” Our hostess has talked about the importance of doing the social-maintenance work of open-source, and why that’s both undervalued and critical. So are many of the “non-hardcore” (and yes, those are sneer quotes) jobs in software.
One — and it’s only one — of the ways to encourage women into the OS community is to celebrate the geeky but non-hardcore jobs that they’re already doing, in both their professional and personal lives. As a 1970’s-era feminist, I’ve seen what happens with the “women ought to have the same jobs as men” philosophy taken to its extreme, and I don’t like it. It led to women who loved the traditional women’s jobs being slighted.
I left coding for complicated reasons, some of them having to do with internalized sexism, the whole “women undervalue their expertise” thing. On the other hand, the instant I discovered the profession of tech writing existed I was enchanted, and I’ve been doing it joyfully ever since. I’ve worked on OS teams. I’ve worked on compiler teams. I’ve worked on class library teams. I’ve worked on stuff so secret that I’ll never be able to talk about it. And ain’t I a geek?
I find your term “women’s work” problematic and essentialist, even if you are defining it in special way that is different from most people’s connotations of what “women’s work” means. Sure, “non-hardcore geek” work is important, but redefining core geekiness does not solve the problem of men stereotyping women as less competent in coding.
Maybe we have different views about what “the problem” is. Your problem is (perhaps) moving goalposts of who counts as a geek to exclude women, but my problem is that women are perceived to have a narrower range of competencies than men.
I was attempting to use “women’s work” as shorthand. I agree with you that the term is loaded and essentialist and a bad choice; its current meaning is too well established to be redefined for the purposes of argument. My bad.
However, “work traditionally dominated by women” is long and clunky, and I should be grateful for a shorter substitute.
I don’t think your problem and my problem are exclusive. It is both true that women are not accepted as peers in, and are in fact discouraged from, the high-prestige parts of the community and that the valuable jobs that women have traditionally done in the geek community are undervalued.
I do think that creating a culture in which all aspects of tech work are celebrated is a component in encouraging women to access work which has traditionally been male-dominated. On the OTW’s Archive of Our Own project, one of the things that I value the most is that the contribution of every discipline is recognised and celebrated. Because there’s an atmosphere of mutual respect between coders / testers / support staff / etc they work more closely together, and I think that this makes the divide between one discipline and another smaller. We’ve had people move from testing and docs to coding (and we’ve also have people move in the opposite direction), and I think their readiness to do that was closely connected to the fact that coders aren’t seen as some separate and special breed of people. Obviously, this project has some specific aspects which make it more women-friendly anyway (the fact that it’s a 97% female team, for a start!) but I do think that celebrating the work on non-coders can help to break down the barriers between disciplines.
As a consultant (for major financial firms with my focus being in capital markets technology) I work with many groups – infrastructure engineers, software developers, business analysts, trainers, end users … I’ve noticed that the gender ratio changes in two ways – as one gets closer and closer to “the code,” the number of women decreases. Then there are more and more of us as one goes out towards the product (especially customer service and training) and then an abrupt drop off for the actual end users (the traders themselves). Essentially, I’m in the middle of two different “anti-female” worlds – hard tech and debt/equity trading.
Here in the middle though, I am remarkably surprised with just how many women there are all over the place and how diverse their roles are. The team I currently work on (granted, right now we are doing BA work and documentation) is 3 women and 2 men, under a female PM. And every single one of us is of a different ethnicity. Personally, I only feel the whole “geek cred” pressure because of the non-professional circles in which I travel, where being a hardcore geek is a badge of honor. At work, the tech folks doing this “for their job.” They’re an older set and I rarely get the feeling they think I’m “wouldn’t understand” or “shouldn’t be there.”
Just my experience.
PS – As for the scripted/compiled thing, I can say that I started coding so I could make things for the web, but later got into learning C because it seemed like “something I should know” and now am coding for my Arduino. My C is passable, my mySQL/PHP is pretty good (IMO). Since I didn’t go to school for CS, what I learn is simply dependent on what I want to make. So can we call scripted languages “gateway coding?” ;)
Awesome rant, Jonquil!
I’m really wary of getting involved in geek prestige stuff, because in my experience once you scratch the surface it is pure and simple sexism. You can’t improve the perception of stuff women are good at, because that’s the whole aim of geek prestige, to shore up (some) male egos.
Most geeks (of the sexist variety) value stuff that is unintelligible to non-geeks – that’s what makes it prestigious and appear “hard”. In actual fact, writing code that can be used easily and intuitively by non-geeks is a much bigger challenge, just like writing good documentation is a much bigger challenge than the (mainstream) geek culture acknowledges. It’s good old-fashioned “secret clubhouse” mentality, and at least I am not willing to spend my time trying to change men who think like that.
In my mind, the push to increase respect for non-coding roles is twofold; 1) by de-emphasizing the code writing we can start to appreciate the many important and skilled tasks within our community that are already being done and 2) by highlighting those roles we can perhaps attract new people to the free software community and maybe while we’re at it, getting some of them close enough to the process of code-writing so they can see that it isn’t “magic.” *
* I don’t mean to say that coding isn’t hard, but from the outside it can look pretty intimidating.