We've had a lot of requests from the community on some general statistics for the program, and we'll have more posts in the coming weeks on various interesting statistical tidbits about our students of yesteryear and today.
First, the pretty good news. Consistently between 2005 and 2007, approximately 4% of SoCers are women. While we'd like to see that number increase over time, we're still quite happy that it is double the usual percentage of women reported in the overall open source developer population. We're also proud to support some other efforts to encourage women to participate in open source development, such as the GNOME Women's Summer Outreach Program.
The very good news is that our community is passionately invested in getting more women into open source development. Each time I talk to members of our mentoring organizations, I consistently hear enthusiasm around having female students working with them. And where organizations haven't worked with women in the program, they are actively seeking ideas on how to attract more women to their projects.
We'd like to hear more from the community on this topic: how can more women be attracted to open source development? Or, to put it a different way, how can we ensure we're shipping way more women's t-shirts next year?
You'll notice that comments are now on for the blog, and we look forward to hearing from you. Please bear in mind that all comments should be respectful and constructive. (In other words, don't make us use that delete button.)
Google Summer of Code Blog
News on Google' s program to encourage
student participation in open source development.
Friday, May 11, 2007
Women in Google Summer of Code
Subscribe to:
Post Comments (Atom)
23 comments:
Well, first we need to see exactly where the problem lies. Were there so few women applying for SoC to start with? Or did they got more proposals rejected than men? See which organizations women have applied the most, and which have accepted more women, and try to infer the cause? Check what kind of stuff they prefer coding, and what organizations could represent them better. Etc. Work with the data you got...
If there were few women applying, I don't think there is much more you can do than trying to attract more people to apply in general.
Blacksheep I can pretty well guess you are not a woman, although I appreciate your well thought out post.
My suggestion is to form some sort of group working on widgets which have value in "social" applications, e.g., something which, for each of the users in a chat room, displays a little clock giving the user's "local time."
Not that I am enamored with this particular example. Just hoping some of the women who are already involved can come up with better ideas.
teresa: there's a few issues with that. One is that SoC is based around existing projects, rather than forming new groups. Another is that there are already a number of "social" applications in the project, the one I work on being a good example... and yet we received a grand total of zero female applicants (that I recall, anyway. I don't pay much attention to the names attached to proposals until I start getting to know the person).
I think there's some deeper issue here than mere project choice.
There's nothing I can think of about coding that males, by any stereotype you choose, would be better at than females.
It's worth noting that women are already graduating colleg at higher rates than men.
There's no way to look at these numbers but to conclude something, somewhere is wrong with what we're doing, as open source communities.
Four percent is not good-- we are new people to open source coding, our group should be at least half female to start eroding the imbalance.
The Drupal projects have I think just one woman out of twenty– which technically means we're raising the average, but that's no comfort.
I'm normally more concerned about class inequality than sex inequality, which after all isn't compounded across generations the same way. And I still think there needs to be Google Summer of Code type outreach to people who in particular *don't* go to college. Open source provides degree-free opportunities to make a living and contribute to society like few things I can think of. (It may similarly help close the gender pay gap.)
We need outreach, and perhaps a more encouraging culture for those who do get involved. (Drupal does do well at this, with women coders among the leaders -- add1sun and webchick to name some names.)
The importance of broadening involvement of disadvantaged groups and women isn't just for their sake– it's important to open source: to have more people; to have people with different needs in a system that develops software useful to communities in part by meeting ones own needs; and to reinforce the best parts of a sharing-based ecology from groups historically better at sharing.
(Where I was trying to go with that is that we need a general "Yes, you can" approach to involving people in open source -- there are significant and strange psychological barriers to programming for anyone, for instance I don't even imagine doing desktop applications even though I'm sure I could -- with a particular attempt to contact women and poor and any group we're not reaching.)
The best study to date of the gender imbalance in computing as a whole, and how it can be leveled out, is Margolis and Fisher's Unlocking the Clubhouse. They identify several factors, some of which kick in early (families and schools tend to view computers as "boy's toys"), and some of which act later on (male students' bragging about how much they already know can give female students a false impression of their standing). M&F's work at CMU shows that when these factors are addressed, gender balance improves dramatically; this is in stark contrast with the continuing decline in the number of women pursuing computer science degrees in the United States.
A more recent book that's well worth reading is Ceci & Williams' Why Aren't More Women in Science? (reviewed here). This collection was put together in the wake of Larry Summers' comments at Harvard, and presents arguments from both sides of the nature vs. nurture debate on female participation in STEM (science, technology, engineering, and medicine). It's interesting to see how often the same data is quoted on both sides of the argument...
Finally, I'd be very interested to see the data behind the claim that 2% of participants in open source are female. When Michelle Levesque and I looked at this in 2004, we came up with a much lower estimate. No matter which is right, it's much (much) lower than the 28% often quoted for the industry as a whole (based on US data circa 2000). I think that's what we should be comparing outselves to.
@ blacksheep: We certainly had fewer female applicants than male applicants. We're hoping to be able to attract more female applicants in hopes that we'll have more women accepted into the program.
@ gvwilson: Agreed that Unlocking the Clubhouse is a great look at the gender dynamics of computing.
I'd say we have consensus that we need more women in open source. Does anyone have suggestions for improving this trend or specific thoughts on outreach?
I recently did a bunch of research into this very topic for a talk I gave at Flourish! conference on the topic. There are slides available, if you're interested (which encompass a lot of fantastic input and feedback from the LinuxChix), but the short version is...
1) IMO, while there are certainly social reasons why women don't get involved in open source (stereotypes, etc.), those reasons are also generally true of *all* women getting into IT. While women are involved in IT in general at a much lower rate than men, the open source rate is, quite frankly, abysmal. So we (the open source community) need to look at what we're doing to either actively or passively exclude women. But in a way, it's good that it's stuff we're doing, because we have a lot more power to change and improve those things. ;)
2) What are some of these things that are barring women (and new contributors in general) from getting involved in open source? Tolerating "poisonous people" who tell sexist/racist/homophobic jokes and put down others (hint: this crap turns away men as well, or the smart ones anyway :P). Telling people to RTFM instead of taking 30 seconds to point the way. Placing up unnecessary barriers in front of people before they can get involved.
3) What can we do? We can actively promote an environment of mutual respect. We can make it as easy as possible for someone who wants to help to do so: write better documentation, create mentorship programs (like SoC, Drupal Dojo), etc. We can work to dispel the commonly-held myth that you have to be a super genius to participate in open source. And if we are women already involved in open source, we should make ourselves as visible as possible, to help encourage others to come join the fun!
Most efforts that go into making it easier for *new contributors* to get involved will also extend to making it easier for *women* to get involved. Drupal has seen this with Drupal Dojo, which is real-time, interactive peer-tutoring over Skype/VNC/IRC. I know of at least a few new women who gained the confidence to try their hand at contributing to Drupal after spending some time here (and one super-star in particular who is co-teaching classes, creating videos for other people to help them get involved, etc.).
And personally, I don't think I'd be involved in open source at all if not for SoC. It poked a tiny hole in that seemingly impenetrable "you must be THIS smart" wall surrounding open source development. I applied because I figured, "Well. If they know I'm a student, they can't possibly expect me to know *everything* yet, right?" After SoC allowed me to cross over the "open source contributor" threshold, it made me feel kind of cheated that I hadn't done so MUCH sooner. But it's ok, I'm making up for lost time now by going bananas and helping out everywhere I possibly can. ;)
Sorry, that got a lot longer than I thought. :P I don't know if it even helps any, but there you go. ;)
In my current hardware class there are 12 students.. 2 of us are women. Most women I have met find computers boring and not desired. Others who actually are programmers or in other areas of the tech field have no interest in being apart of any community. They just want to do their 9 to 5 and go home. I think the increase is a good thing and will only get better.
I hang out on irc.freenode.net in alot of help channels. Yes, RTFM or triggers and pointing the link is annoying and maybe a bit confusing to anyone who is new and trying to learn. I think that the opensource community just doesn't offer a lot of things that women are looking to use. Unless you are already a computer geek willing to broaden your horizon. A lot of the women I have met have never heard of sf.net or open source. I think spreading the word about open source is needed for both women and classes to make things more equal.
As far as all the negative stuff. A lot of people in general do not want to deal with the sexism/racism jokes or not. So they rather just be on their own than deal with a community who ignores it. Sadly people take freedom of speech to far.
@ webchickinator: Glad to hear that SoC is effective in lowering the barrier to entry for new developers.
@ danyelle gragsone: Well said. If a project is clearly unwelcoming to a new developer, they've just lost out on their opportunity to get help from a new contributor. I think the situations you describe - racism, sexism - are likely to drive away both female and male contributors, though women likely moreso.
Next we do something about the northern hemisphere bias? Every student who applied for our projects from up north. I'd love to encourage New Zealand women to participate in Winter of Code, but not in the middle of school term.
@ taniwha the wally: The timing for the program may be suboptimal for folks in the southern hemisphere, we still have many students participating in SoC who are based there. I'm not sure that the timing is a barrier to entry for women specifically, but your point is noted.
I agree that computers and technology just don't attract women that much. Additionally, I'd guess that open source in particular appeals to even fewer women because it's more about ideology than development methodology. Certainly, open source was started to remove the political baggage from free software but in current use, it's basically the same thing but with a less ambiguous wording.
I think that's the main problem. Young men tend to be more idealistic and rebellious than women, so it gives a young male geek the feeling that he can do his part to change the world (disillusionment will likely come later). Women are more practical, so if you want to attract more women to the open source movement, then you'll have to think of some practical reason for them to join (and calling it a movement doesn't help any), which may be hard in a community filled with male idealism.
@ zooplah: I must respectfully disagree that young women are less idealistic than young men. I certainly was a highly idealistic young woman and remain so long after my college years have concluded.
I think that many see the open source realm as the perfect way to 'scratch their own itch' while doing something noble and of benefit to others. I think women may be more inspired by the community aspect of open source projects than the pure coding aspects.
Do others have thoughts on this assertion? If people think the community aspects are more appealing to women, what could (or should) be done to promote the community side of open source to a larger audience?
IMHO, as women ...
I think software production need women touch because not all of user are men :D
Involving women in OSS makes it different things. Women and men are created for working together.
Student of Joomla SoC
@ nur aini: Agreed - it is disconcerting that women are half the population of the planet, but that so few of them are contributing to the conversation that is technology development.
I just looked at the speakers' list for RailsConf 2007, and saw 71 male and 5 female speakers (6.5% female); of those who had photos posted only one was non-white. I'd be interested in hearing if this is representative of the conference as a whole.
http://pyre.third-bit.com/blog/archives/942.html
@ sheeri: Since you don't see the dearth of female role models for women in technology as a barrier to entry for yourself, do you think there are other ways to encourage women to participate?
Thanks for being one of our mentors!
ah well, going through all these differnt postings which I just skimmed through, I'd say first of all, most of us would need a change of mindset and prepare ourselves how we work on open source. I guess having social apps is great, and would probably improve women participating in blogs.
However, on a larger scale of things, what I think is that in order to get more women involved at the same, you need to first have some form of discussions with women who are already working in open source projects. Do a greater exposé on all these ppl and get them involved in a community project. Try to get them participate in their own womenly dynamics of doing things, so that they can work in parallel with existing groups which are male. Best if someone could start a movement. Hopefully with less feminist slant though as that would tip things the wrong way and upset many people. That of course being said would be best being aided and facillitated by males who are more comfortable working with females.
With all that being said, we still have to accept the fact that open source is still predominantly a male thing.
This is not because we do not let females in, but just that the very nature of open source is started by males, and involves a lot of pioneering work which very few females are open or accustomed to.
@ tulipsoss: Some of your suggestions are good, and for those looking for more detail on women's open source groups, I'd recommend checking out LinuxChix. It's not just for women interested in Linux, but in all types of free software. The group also offers tutorials and classes by email.
However, I strongly disagree with you that women are not used to doing "pioneering work," in open source or other fields. Neither sex is predisposed to innovation.
Hmm, I might not have worded that correctly, wasn't really suggesting that women are any different.
Anyway, the basic point I was trying to push was that there are statistically there are probably 5 to 10 times more men than women in computer science and mathematics circles and being in this kind of fields requires alot of thinking not many females are accustomed to, unless its due to necessity. Cos for most females, innovation happens because of a dire, inherent basic needs that needs to be fulfilled and they have to innovate in order to survive. Well, in short, I'm just trying to focus on the point that a guys mind works and functions differently from a gals mind and the pioneering work i'm talking about are the kind that is a normal part of a guys mentality, while for women, it comes and manifests in slightly different forms.
Anyway, I'm not sure if you heard about Mohd Yunus and the foundation of the Grameen Bank in Bangladesh, but in that aspect, its sorta like an "open source" culture and they have a system where they give back to the community and in the end the bank is owned by the people who borrow, with most people being women. I find it really interesting on how the bank ended up serving women in microfinancing in order to achieve what the founder planned.
Maybe we could learn a lesson or two from there? You can view the webcast of the Khazanah Global Lectures with the talk by Dr. Mohd Yunus titled "Building Tommorow's Minds today", which was conducted live with 8 Malaysian Universities at http://oss.mmu.edu.my/videos/khazanah/MMU%20Live%20Streaming%20Khazanah%20Global%20Lecture%20with%20Dr.%20Muhammad%20Yunus%20Building%20Tomorrow%27s%20Minds%20Today.wmv to get a better idea on what I was talking about.
Post a Comment