The numbers of women in the field of computer science are dismal and even worse in the arena of FLOSS; a widely-cited study showed that only 1.5% of FLOSS practitioners were women (as compared with 28% in proprietary software).[1] As scholars of the ethics and politics of computing, we are committed to generating data and providing theoretical insight into the formal barriers, subtle and implicit biases, and missed opportunities that keep women from fully participating in technological production. Due to our expertise and commitment to these issues, we feel the need to respond to a recent New Media & Society article, “‘Patches don’t have gender’: What is not open in open source software”, by Dawn Nafus. [2] We are at once unconvinced by core arguments and troubled by the distortion of an empirically complex set of phenomena surrounding women’s participation in FLOSS, and we see an opportunity to raise awareness and spark discussion.
The article diagnoses a problem hard to disagree with: “While open source software development promises a fairer, more democratic model of software production often compared to a gift economy, it also is far more male dominated than other forms of software production” (p. 669). Where we start to take issue is her core claim: gender is a topic so taboo that for many in FLOSS, “women’s absence posed fewer problems than the method to change it” (p. 674).
Complexity in the record: Debian Women
Debian, the largest free software project in the world, is a primary example for Nafus, as she evaluates (male) support for its women-based initiative, Debian Women. According to Nafus, community response was at best tepid and at worst met with suspicion and controversy. She quotes a blog post where one person commented, “I think the whole idea of ‘Debian Women’ is flawed. All it does is give / reiterate to people the idea that women are somehow different to men when it comes to computers and should be treated differently.” (p. 673).
Yet to rely on this blog thread in near isolation is to overstate the case. If one includes the people involved in the chartering of Debian Women, the picture starts to look different, very quickly: The idea for the project was hatched at the annual Debian conference in 2004 held in Porto Alegre, Brazil. Spearheaded by the efforts of Erinn Clark, the few women attending used the time and energy afforded by an in-person meeting to initiate and organize the Debian Women Project, a Web site portal and IRC mailing list to encourage female participation by visibly demonstrating the presence of women in the largely male project. During the course of the conference, Erinn Clark gave a presentation “Women in Debian” to a packed room of mostly male developers, the great majority who responded with excitement and support (one of the authors of this piece, Gabriella Coleman, was in attendance at the conference and talk).
Indeed, following the conference, one female Debian developer, Amaya Rodrigo, posted a bug report calling for a Debian Women’s mailing list, explaining the rationale in the following way:
From: Amaya Rodrigo Sastre
To: Debian Bug Tracking System <submit@bugs.debian.org>
Subject: Please create debian-women mailing list
Date: Tue, 01 Jun 2004 22:12:30 +020
Package:lists.debian.orgSeverity: normalOut of a Debconf4 workshop the need has arisen for a mailing list oriented to debating and coordinating the different ways to get a larger female userbase. Thanks for your time :-).[3]
We find it politically significant that Rodrigo turned to a technical platform to diagnose a “social” problem; this belies Nafus’ repeated claim that FLOSS developers hold technology to be orthogonal to the social (p. 674; passim). Contrary to what one would predict using this optic, which assumes such incursion of gender into the technical domain would be met with skepticism by the mostly male developer cohort, comments left on the bug report were, in fact, overwhelming supportive. Two examples are:
i support the idea. about time IMO. oh, and i would have upped the severity [of the bug report] a bit i think ;). s/
I wholeheartedly second this idea — having the women of Debian work on corrupting my wife and turn her into as devoted of a Debian nut as I am is a project definitely worthy of its own mailing list![4]
Gender advocacy in FLOSS is ongoing
Moving beyond the historical example of Debian Women reveals other contours of women’s participation in FLOSS. Empirical study indicates that while some of what Nafus reports (e.g. extreme sexism) may certainly be found in FLOSS communities, its presence cannot be assumed in all communities to the same degree. FLOSS projects are not monolithic, nor are FLOSS project responses to gender imbalance all identical. For example, if Debian Women was careful not to exclude men, Ruby and Python groups “for women and their friends” have glossed this slightly differently. They paint men as potential allies, but men are unwelcome in certain workshops unless accompanied by women; they may be invited as “guests” but they are to take a back seat. “Women”, the primary constituency for these are workshops, are explicitly defined as woman-identified, including trans- and queer women.[5]
Thus the notion that gender ought to be “irrelevant” is far from being a universal refrain. Indeed, there is a wealth of advocacy and activism within FLOSS communities to address “diversity”, especially gender. LinuxChix was formed over a decade ago, and Debian Women was founded eight years ago. A raft of newer initiatives including RailsBridge in the Ruby community and PyLadies and PyStar in Python have since formed. Most of these groups are absent from Nafus’ discussion, as is any substantial exploration of the perspectives and experiences of women developers, a surprising omission given the topic of the article.
The point is, many members of FLOSS communities would likely disagree with the picture painted by Nafus: there is growing sense within some projects that gender and diversity are worthy topics for discussion and that blatantly sexist comments are not to be tolerated. This is not to say that no one in FLOSS communities argues, as Nafus claims, that gender is “artificial” and therefore ought to be “irrelevant”. She is certainly correct that a good way to set off an argument in some contexts is to raise the issue of gender; a typical instance was the kerfuffle over a reference Richard Stallman made to “Emacs virgins” in a 2009 presentation, which prompted outrage from some quarters about the potential for this remark to alienate women, and responses from others arguing that singling out women as deserving protection from this sort of harm, i.e. potential verbal offense, was itself sexist.[6]
Yet notably, in the fracas over Stallman’s comment, male practitioners were among those attempting to police this behavior and establish that such remarks were inappropriate. At the same time, not all comments raising gender as a topic for consideration set off heated discussion. Both of these facts indicate that some practitioners—both men and women—do not question that showing support for women in tech fields should be a goal of FLOSS practitioners.[7]
Nafus is also correct that some women are distressed by being singled out to repeatedly comment on their gender. Some people indeed feel that this is tiresome and unnecessary because gender should be irrelevant. On the other hand, others feel something rather more elaborate, wishing to acknowledge the issue of participation by women but being ambivalent (or simply fatigued) about being spokespeople for or ambassadors of their gender. Furthermore, while women may experience ambivalence regarding their visibility, some have concluded that this is a fair price to pay to change the constitution and practices of the groups in which they participate: in the words of one hackerspace participant, “Emma Jane Hogbin’s Unicorn Law[8] states that ‘If you are a woman in Open Source, you will eventually give a talk about being a woman in Open Source.’ Yeah, well. The thought here is having more women in visible roles in the workplace and the culture normalizes their presence and provides role models. So go ahead and volunteer for committees and panels.”[9]
Beyond “women”: Theorizing gender and technology
Furthermore, a framework incorporating gender troubles that arise beyond the issue of women in FLOSS can provide greater analytical yield. Nafus writes, “This [academic] literature shows how programming cultures sustain certain forms of masculinity which make women concerned about being ‘unfeminine’ in their connection to technology or ‘too feminine’ by attracting unwanted male attention” (p. 671). The notion that “the gender problem” in FLOSS centers on a double bind in which women have to be careful not to be “too feminine” while also not wishing to feel “unfeminine” in their pleasures in technology collapses an ultimately more complicated set of issues. One major pitfall of this framing is that it serves to reproduce heteronormativity, in which there is no possibility for women to possess or express anything but femininity.[10] Yet in reality, researchers acknowledge multiple femininities and masculinities and the reality of complex dynamics vis-à-vis gender and technology, including commonplace existence of men exhibiting “non-masculine” values such as sharing, humility, and reciprocity in FLOSS.[11]
A strict binary understanding of the gender-technology relation undercuts the rich complexity of technology as a site for negotiating and constructing gender; empirical evidence suggests that technical practice is a site where some participants experiment with gender crossing and liminality.[12] Moreover, hacker and FLOSS communities contain salient non-binary, queer, and trans contingents. In interviews conducted by Christina Dunbar-Hester within the past year, some participants speculated that for hackers, gender, like technology, is just one more domain to question “why things are this way as opposed to that way” and “hack on”, i.e. a site to display curiosity and assert “maker” agency. Others suggest software development is an occupation that may appeal to people who do not “pass” or conform to gender norms, willfully or otherwise, because corporeal presence at a workplace may be less strictly required. There is certainly no single or simple explanation for this phenomenon, but its existence—and indeed prominence—should be noted by scholars of the gender-technology relation more generally and hacking and FLOSS in particular.
Conclusion
In sum, in setting the historical record straight and calling for a more expansive theoretical understanding of gender in FLOSS, our motivation here is not to serve as apologists for FLOSS communities. There are many problems to address and redress. The conversations occurring there about gender, sexism, and inclusion/exclusion are urgently needed, and often complicated and difficult. However, neither scholars nor practitioners are well-served by flattening the heterogeneity of conversations within FLOSS communities about gender. In this case, misrepresentations of sociological and historical dynamics are especially worrisome for they overlook the rich range of initiatives that community members—men and women, queer and trans people—are taking to redress gender imbalance. In other words, the problem is not primarily that gender is off-topic within FLOSS. Indeed, a lively conversation about gender is flourishing on project mailing lists, on blogs and at conferences; these conversations and interventions are an essential starting point to grapple with the limits and possibilities of strategic efforts to engender change in FLOSS.
Works cited
Coleman, E. Gabriella. Coding Freedom: The Ethics and Aesthetics of Hacking. Princeton: Princeton University Press, 2012.
Dunbar-Hester, Christina. “Geeks, Meta-Geeks, and Gender Trouble: Activism, Identity, and Low-Power FM Radio.” Social Studies of Science 38.2 (2008): 201-232.
Landström, Catharina. “Queering feminist technology studies.” Feminist Theory 8.1 (2007): 7-26.
Nafus, Dawn. “‘Patches don’t have gender’: What is not open in open source software.” New Media & Society 14.4 (2012): 669-683.
Pinch, Trevor and Frank Trocco. Analog Days: The Invention and Impact of the Moog Synthesizer. Cambridge: Harvard University Press, 2002.
Rommes, E. W. M. “Heteronormativity Revisited; Teenagers Educational Choices, Sexuality and Soaps.” In S. Booth, S. Goodman & G. Kirkup (Eds.), /Gender, IT and sites of learning/. IG Publisher, 2010.
Streeter, Thomas. The Net Effect: Romanticism, Capitalism, and the Internet. New York: New York University Press, 2011.
Wajcman, Judy. Feminism Confronts Technology. University Park, PA: Pennsylvania State University Press, 1991.
[1] Ghosh, R. A.; Glott, R.; Krieger, B.; Robles, G. 2002. “Free/Libre and Open Source Software: Survey and Study. Part IV: Survey of Developers.” Maastricht: International Institute of Infonomics /Merit. Though these statistics are a decade old, there is little reason to believe the situation has improved dramatically since this study.
[2] Nafus and colleagues conducted quantitative and qualitative research though this piece focuses on qualitative. Newer numbers than the FLOSSPOLS study are needed and we look forward to seeing an update to the earlier report (“Free/Libre and Open Source Software: Policy Support”, FLOSSPOLS Deliverable D 17 Gender: Policy Recommendations, UCAM University of Cambridge, authors Nafus, Leach, Krieger, 2006).
[3] Debian Bug report logs, http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=252171, June 2004
[4] Debian Bug report logs, http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=252171, June 2004
[5] Boston Python User Group, http://meetup.bostonpython.com/events/42610202/, December 2011
[6] See post and comments at http://opensourcetogo.blogspot.com/2009/07/emailing-richard-stallman.html, accessed 4/1/12.
[7] See also “The tech world needs a “white ribbon” project: Support for women in tech-related fields needs to be made visible”, http://peak5390.wordpress.com/2012/03/23/the-tech-world-needs-a-white-ribbon-project/, March 2012
[8] See http://geekfeminism.wikia.com/wiki/Unicorn_Law
[9] “11 To-Do’s for Women in Tech,” http://www.thehacktory.org/?p=1911, December 2011
[10] Landström 2007: 11-12; see also Rommes 2010.
[11] Coleman 2012; see also Wajcman 1991.
[12] Dunbar-Hester 2008: 215-218; Pinch and Trocco 2002: 138; Streeter 2011: 202, f.n. 16.