The founder gap: Why we need more women in open source

No readers like this yet.
Lots of people in a crowd.

Opensource.com

Look at the founders of any Internet startup, and you're almost certain to find an open source expert among them. Take Google, the biggest Internet success story of them all. Co-founders Larry Page and Sergey Brin built and ran their brand-new web indexing tool on servers running the open source operating system Linux—and still do, 13 years and several billion dollars in profits later. Pick an Internet startup—Facebook, Zynga, Groupon—and chances are you'll find open source software running on its servers (and an open source-savvy founder in the boardroom).

At the same time, women make up an estimated 2% of the open source community, far lower than the percentage of women in computing overall, estimated at around 20%. Is it any wonder that women founders are so rare in Internet-related startups, when many of the founders come from a population that is 98% male?

If we want to increase the number of women founding Internet startups, it is imperative that we close the "gender gap" in open source software.

And we do want more women founding startups: Study after study finds that startups (and corporations in general) with women in top management spend capital more wisely, succeed more often, and make more money in the long-term. Products and services built by more diverse teams are better. And it’s just plain fair: the billions in personal wealth that end up in founders’ pockets shouldn’t be limited to "white, male, nerds whove dropped out of Harvard or Stanford."

What am I doing about it? Well, I quit my job as an open source developer and co-founded the Ada Initiative with Mary Gardiner, also an open source developer and feminist activist. Named after Ada Lovelace, the first computer programmer, the Ada Initiative is a non-profit dedicated to increasing the participation of women in open source software, Wikipedia, and other areas of open technology and culture. We work through education: teaching workshops, writing guides, giving free consulting, and more. We teach women skills for succeeding in open tech/culture, and men how to support women.

Want to help? Support the Ada Initiative through our current fundraising campaign and learn more about our work.

Originally posted on The Ada Initiative blog. Reposted under Creative Commons.

User profile image.
Valerie Aurora is the Executive Director of the Ada Initiative and has over ten years of experience as both a Linux kernel developer and as an advocate for women in open source. She worked as an operating systems developer for several leading open source companies, including Red Hat, IBM, and Sun Microsystems.

20 Comments

[Removed] We need good programmers, not some politically correct <em>BS [edited]</em>, not incompetent people with or without [Removed].

Do the world a favor, and [Removed].

<em>Editor's update</em> We removed some vulgar language from this comment that violates our <a href="https://opensource.com/legal">terms of use</a>. This is not acceptable behavior on opensource.com and it will not be tolerated. However, we do appreciate the follow-up support from the community which is why we choose to edit the comment at this time instead of removing it completely.

AC: We can't get half of the potential good programmers because they don't want to put up with the dickhead comments like your own.

Cheers AC, gave me a reason to donate again.

You've made up my mind to contribute. Thanks for proving their point for them so elegantly...moron.

Thanks AC for showing how much the Ada Initiative needs my money to combat ignorance and mindless agression like yours.

AC used a fake email. Which highlights his cowardness. Thanks for the community support. Vulgarity removed. Not appropriate on opensource.com.

Jason

Thank you Valerie, for bringing awareness to this issue.

Awareness is what this article is all about. And it's what the goal should be right now for the men and women who understand what's going on. Because, for any field or industry to fully live up to its potential, and not miss great opportunities for advancement and innovation, both men and women should not only be invited to the party, but made to feel welcome there too.

Good stuff.

While I would welcome more women in IT and open source in particular, I do have some concerns about the efforts being made. Working at a university, I have seen and contributed to many programs to bring women into STEM fields. I don't see a lot of progress, even at places where these programs promote women in STEM while excluding men. (which may or may not really be a problem) A few years ago, one of our graduate students did her M.S. project on this very subject. I found several facts from her literature review puzzling.

The number one reason given by women for enter the IT field was the prospect of a well paying job. The number one reason for leaving the IT field (or avoiding) was "I want to work with people, not computers all day". In addition, a large percentage of women leave the IT field after 5 years and never return.

So my questions start with, are we pressuring women into a field that they won't be happy with? Will women ever represent more that a small fraction of people in this field? Do women fgure out that money isn't everything, causing them to leave the field to pursue another dream? Nursing, in comparison, is 95% women, 5% men. Yet no one is really surprised or troubled by this. Maybe IT is just a field that men are more attuned with than the average woman.

Personally, I would like to see a change to the efforts of recruiting a gender into a field. I would like to see a national movement to provide all high school students, men and women, with career experiences throughout high school. Many students have no idea what they want to do after high school because they lack experience. I know I personally was pressured into a field I later found I hated (engineering) because of the prospect of a good job. While I do well in IT, my job views are no longer based on money. IT is something I love to do.

I'm going to guess that role models in high school and college matter a great deal. I only had 2 in college. One was Don Greenberg, who wasn't really a CS guy but rather an Architecture guy, who did early work in 3D graphics. His approach was entirely hands-on, compared to a CS dept. that was mostly theoretical mumbo jumbo back then. The other was Devika Subramanian, whose class I happened to take as a senior because she was just so interesting and human in her sales pitch for it. The class was about databases, not a subject I had any inherent interest in.

The rest of the CS faculty were complete duds. They were awkward, and talked about a lot of stuff that doesn't matter. I ended up majoring in Sociocultural Anthropology, where smaller groups of people would have meaningful human conversations / debates about important things. This despite my natural aptitude for mathematics, physics, and computer science. I liked the human beings and didn't like the geeks, even being half-geek myself. Any Anthro geek was far more personable and worth speaking to than a CS geek, even the squirreliest of the Anthro professors.

After college I knew that 3D graphics was what I wanted to do, so I taught it to myself. I doubt most people do that.

Do something about the culture of CS depts. being totally repulsive to "more normal" human beings and it's possible that more women will be interested in taking classes. In my case I think my Nature won out, because Nurture was definitely pushing me away from CS for a time.

Gary,

You bring an interesting point, that goes to the core of the issue: the reality or perception that work in IT is disconnected from human interaction.

Seen from the surface, and when observed over short periods of time, it is common to get the perception that the IT field is devoid of human connections.

However, the more one works on open source and on software in general, the more evident it becomes that the real issues in our communities and projects are not tecnological problems but "people's problems".

Many projects languish due to social friction among the contributors, and lack of skills on managing human motivations and on appreciating the type of rewards that are valued in a gift economy. (In plain words, there is a lot of "jerk" behavior among us, which sometimes is more tolerated among males, but that still is equally damaging to the communities).

What is true in IT in general, becomes more accute in an open source environment where a lot of the work is done on a volunteer basis, and a good deal of motivation arises from the pursuit of meaningful-work, connection to something larger than ourselves, satisfying interacions, and development of a sense of belonging.

The consequence of disregarding the importance of a balanced composition in the social structure of our open source communities, is that when projects reach the maturity and dimension when it is evident that people's issues are becoming more prevalent than the day-to-day dealing with the buzzword technology of the day, (typically when they have more than 20 contributors and more than 5 years of history) they find themselves devoid the gender diversity (and age, and race diversity) that would have empowered them to overcome the social structure challenges, to grow further, and to turn into a powerful and productive social organization.

The advocacy for diversity (gender, age, race,...) is usually presented in terms of social justice for the underrepresented groups. In our case, it is my opinion that it is urgent to pursue diversity for the purpose of leading our open source communities to reach their full potential.

Not having women evenly represented as active participants of open source communities, leaves our communities in a vulnerable situation where we can not reach the scales of hundreds or thousands of participants, much less the level of hundreds of thousands, where the benefits of the network effects of openness and free flow of information and knowledge reach the highest returns.

Ohloh reports that, from the 500,000 open source projects they track, about half of them have only two or one contributors, while less than 2% have more than 50 contributors. For each one of those half a million projects, we should ask ourselves: Why didn't this project grew to the scale of the Linux Kernel community or even bigger ? What is it about this nascent community that left them stock in a team of only 2 or maybe 5 people.

I suggest that lack of women participation is one of the main structural reasons why these communitites fail to germinate. I will also suggest that lack of students involvement, and and lack of active engagement of young people (12 year olds to 25 year olds) is a close second reason.

Open source communities should dedicate 30% of their efforts to continuously recruite new participants, and when doing so, we should ensure that these new members are coming from a diverse pool of gender, age and backgrounds.

Much as I'd like to buy the pitch that "women's participation would fix it," I can't. I've never participated in large scale open source. All of my participation has been small scale, those tiny teams that typically end in failure. They fail because (1) people don't have any idea what it takes to make a project sustainable over time. They think they've got lots of time to do things when really they don't. Milestones have to be hit or morale slides and real life intervenes. People find out they gotta eat, that they're paying for all of this. (2) People are jerks. (3) People diverge as to what they think is important to pursue. In small projects there isn't enough room for everyone to be a leader, so when it comes to competing ideas of leadership for a project, someone gets pushed out. (4) People don't do the work. (5) People have different strengths and weaknesses for their managerial chemistry. Not everyone can function well in all roles, and people often have poor understanding of people who function differently than they do. This leads to friction and if there's no understanding, explosions.

So, a woman could bring X set of technical and interpersonal skills to the dynamic. It's still a dynamic and having women on board doesn't automatically solve anything. Really, people in open source just have to tough it out for various personal reasons. I am inclined to believe that the root problem is the lack of supply of women in CS.

Oh, and if it wasn't implicitly clear: I think some of us, if we ever get good enough to manage all these issues in any kind of substantive way, start focusing on our own sustainable business models and less on "The Commons." To be that good with people, to get them to actually get a lot of work done, has economic value. Attempting to work with people who diminish one's ability to get things done is a drag.

The problems you mention seem to mainly be management problems. Management sets the rules of conduct for the workplace, makes sure that workers treat each other properly, and so on. Now, open source projects are generally less formal than commercial projects, but good project leadership can still make a big difference.

I see two sides to the issue. On the one hand, it's stupid to throw away the talents and creativity of half the population. I'm sure there's a great deal that women can contribute to open source or any other field of endeavor.

On the other hand, it's silly to assume that the proportion of a given workforce that is composed of members of a particular gender, race, or other group must match the proportion of that group in the overall population. Various factors such as culture may predispose members of a certain group to feel more comfortable with certain types of jobs. For example, men from a culture where machismo is valued would tend to be attracted to more "manly" jobs, perhaps physical labor or some such. Trying to shoehorn people into a particular career just to fit some preconceived notion of equality is not necessarily going to make people happier.

So instead of saying that we need more members of group x in job category y, I'd say that we should try to make all occupations welcoming to anyone who is interested and then let people go for whatever jobs interest them.

Yes those are management problems. My point is that women don't magically have management skills that are appropriate for any given team. They're just as capable of being bad managers or bad participants as anyone else. Most of the small open source projects fail because there's no money in them and people are inexperienced. When people gain experience they tend to move on to more sustainable enterprises, which may or may not have an open source component.

I think you're too quickly concluding that STEM outreach programs don't work.

For example, when I was a high schooler, I was very much on the fence as to whether or not I should get a liberal arts education or a more technical degree. I had the great fortune of earning a scholarship to attend an engineering summer camp for underrepresented minorities (including women and racial minorities) in engineering at the school I eventually ended up going to, Rensselaer Polytechnic Institute. We spent a couple of weeks living in the dorms, working on engineering projects (I remember very clearly the electrical engineering lab where we built our own laser tag guns using breadboards and basic circuit components and then played laser tag around the lab :) ), and also getting lectures from professors and grad students in pretty much every field of engineering. We even took a field trip to IBM's plants in Fishkill and Poughkeepsie NY and saw POWER chips being made.The experiences I had during that summer camp were the push I needed to decide to go to an engineering school and study computer science. I didn't really know all that much about engineering and what it involved before I had that experience.

The majority of my classmates, including myself, ended up pursuing technical degrees, several choosing to go to RPI and graduating successfully and are now working in technical careers.

Another successful example of a STEM outreach program is the GNOME Outreach Program for Women. 40% of the women in the program have stuck around the GNOME project, some becoming mentors for the next round of students themselves (http://adainitiative.org/2012/08/karen-sandler-weve-created-ways-that-we-can-talk-about-it-that-allows-women-to-cope/). This year, the program expanded to include other open source projects, including Fedora, and I'm mentoring a UX intern to work on usability testing for Fedora right now. :) ) https://live.gnome.org/GnomeWomen/OutreachProgram2012

"The number one reason for leaving the IT field (or avoiding) was "I want to work with people, not computers all day"."

Odd. Do you think then women don't belong in white collar, professional desk jobs? Honestly, if you're sitting in a cube in 2013, you're more likely than not working with computers all day, and it's not exactly an option!

"Nursing, in comparison, is 95% women, 5% men. Yet no one is really surprised or troubled by this."

No, no one is surprised by this because traditionally if you were a woman who worked (shocking! Yes, at one point in our culture this was shocking!) the respectable professions were basically nursing, teaching, or secretarial work. This is a historical precedent. Look at how many female doctors there are now, vs. 50 years ago - change in a male-dominated field is possible!

"Maybe IT is just a field that men are more attuned with than the average woman."

So this comment bothers me a lot - I know you didn't mean it to be offensive... but the assumption that women are just not 'built' for certain fields is the very reason why I seriously considered leaving technology as a hobby and going liberal arts when I was trying to decide as a teenager what it is I should do with my life. Who would want to go into a field they've been told (explicitly or culturally) their entire lives they are innately not good at?

I think maybe the issue is at least partly that the IT field is still not fully mature and is still a bit of a 'Wild West' scene, so the types of gender-charged comments and jokes that would never fly in other professional fields and contexts are still far more kosher than they really should be. You can't change the environment you work in, but you can easily change jobs, right? I believe we need a critical mass of women to stick around - I don't think we're there yet - so that we have enough of a percentage of women to actually change this culture. I think the approach the Outreach Program for Women is following is exactly the right one - make a call to women letting them know they specifically are more than welcome to a project, identify those women most interested in/qualified to get involved, and hook them up with a mentor and a support network to make sure they have a positive experience and folks they can trust to go to.

The culture issue is not simply a matter of women needing to have a tougher skin or learn to be less sensitive - some of the comments and behavior I've had to deal with have been pretty outrageous and, at times, terrifying. I would be happy to regale you with horrifying stories over a beer sometime. Or you could take a look at the Geek Feminism wiki's timeline of incidents: http://geekfeminism.wikia.com/wiki/Timeline_of_incidents

"I know I personally was pressured into a field I later found I hated (engineering) because of the prospect of a good job. While I do well in IT, my job views are no longer based on money. IT is something I love to do."

I agree 100% - students should be given the opportunities to experience different career fields and find out what it is they really love doing. Kudos to you to finding what you enjoy :) I love my job in tech as well!

Average *men* aren't suited to STEM either. Analytically gifted men and women are. If you weren't "worthy" material, I doubt you would have gotten into those pre-college programs.

Is there a gender skewing for raw numbers of analytically gifted men and women? I don't know. We all know there's presently a cultural skewing. If we fix that, we have to be prepared for the possibility that the gender distribution will never be 50/50. It may be, it may not be. Policewomen exist now, but they're only <a href=http://www.fbi.gov/about-us/cjis/ucr/crime-in-the-u.s/2011/crime-in-the-u.s.-2011/tables/table_74_full-time_law_enforcement_employees_by_population_group_percent_male_and_female_2011.xls>11.8% of the force</a>. What are the reasons?

What are the skewing effects of a complex economy with millions or even billions of people in it? It's one thing if you're a small country and you need every STEM person you can possibly produce. In a more authoritarian society, maybe girls wouldn't just be trained, they'd be trained forcibly without much choice for their career tracks. Many of the Communist countries used to run people's lives that way. But in a big open economy like the USA's, people can go in all kinds of directions. Whose to say those directions are always going to be 50/50 gender distributed? There's no a priori reason to assume it in a complex system.

"Whose to say those directions are always going to be 50/50 gender distributed? "

This is clearly a straw man argument. There are definitely more women interested in STEM than are actively involved in STEM; whether or not we ever get to 50/50 (in either direction) I think matters less than improving the current status quo which is artifically balanced towards men's favor.

Here's one of them:
http://asherwolf.net/dear-hacker-community-we-need-to-talk/101/

" Policewomen exist now, but they're only 11.8% of the force. What are the reasons?"

This is an interesting question too. There was recently a series of interviews with the new female Congress members, as we've the best female:male gender ratio in US Congress history now. Apparently there was no separate mens' and ladies' room on the house floor, and a ladies room had to be built for the latest session... in fields that are traditionally male, I wonder if some of the issues discouraging women from entering the field revolve around the environment not supporting them even for basics like having an accessible gendered bathroom.

When I said "Maybe IT is just a field that men are more attuned with than the average woman", I was implying desire rather than ability. I fully believe that men and women are equally capable. I am sure there are many career choices that don't appeal to one sex or the other.

As for concluding that STEM outreach isn't having a big impact, I would like to see some results published. Maybe I just haven't seen them, but I wonder, working in education, whether or not what we are doing is working. In the 12 years I have worked at RIT, there has been only minor improvements in the computing programs. I do think that the engineering programs have had better results from what I have seen. Now its possible that the problem lies in the entrenched programs versus newer programs. Our Game Design and Development program has much higher female population than the rest of the college. It just seems to me that there are lots of efforts to push programs and not enough evaluation of results.

You are right in the need to clean up certain attitudes that are prevalent in the industry. That unfortunately I think is a chicken and egg problem. As young men enter the field, the lack of women allows for less that cordial environments where men are expected to just suck it up. But once you have that kind of environment it becomes unwelcoming to women and the problem perpetuates.

" It just seems to me that there are lots of efforts to push programs and not enough evaluation of results."

I think you've hit the nail on the head here. The program that I attended at RPI, the Preface program, I have heard was canceled even though it continued providing good results. However, I don't know how or if they even assessed the results they were getting (presumably they did this at some level as they secured external funding year after year.) The only reason I know it succeeded as well as it did for my classmates in the program is that we kept in touch over the years; the school never contacted us again for follow up information.

Creative Commons LicenseThis work is licensed under a Creative Commons Attribution-Share Alike 3.0 Unported License.