The python community has started a conversation about diversity, with the ultimate goal of creating basically a welcoming statement. It comes out of Kirrily Robert’s keynote at OSCON about women and open source. There is a cool site from the Ruby community called Railsbridge, and one of their guidelines is to “Reach out to individuals and groups who are underrepresented in the community.”
There has been, of course, a lot said about the fact that although women make up 20% of the tech field, they only are approximately 1.5% of open source communities. There have been long standing groups that have tried to address this, and new efforts as well. Some open source communities are more diverse than others. In her keynote, Kirrily talks about two open source projects, Archive of Our Own and Dreamwidth that have a majority of women involved, which is rather unusual.
A short twitter conversation I had with a colleague brought up the issue of whether or not this is just an exercise – will this actually lead to any lasting change? That’s a good question.
Kirrily has a set of really good guidelines for open source communities:
- Recruit diversity
- Say it, mean it
- Tools (Tools are easy)
- Transparency
- Don’t Stare
- Value all contributions
- Call people on their crap
- Pay Attention
As a long time open source user and advocate, even though I am someone who rarely finds people like me in open source projects (i.e other women of color), I’ve always seen the open source movement a potential avenue for the greater involvement of people other than white, straight, young men, because theoretically (this is the important part) one’s involvement in a community is pure meritocracy. But so many open source communities have so far to go when it comes to being welcoming. I’m reminded of sitting in Drupalcon in DC and hearing Dries talk about the “beard length” of the developers. And of course there was the huge brou-ha-ha around a presentation at a recent Ruby conference.
And, of course, there are other factors as well. There are far too few places like The Community Software Lab of Lowell, MA, who’s mission is:
We write, administer and maintain open source software to serve the underserved.
We use and improve the skills of people with underused skills
We work to make hacker sub culture values (transparency, meritocracy and generosity) the values of the entire culture and bring about the post scarcity society.
We work toward our mission by trying to achieve our short term goals transparently and generously while accumulating only necessary wealth.
So what will it take? Will this effort in the python community pan out? I think it’s a great start. I think the first step is definitely a focus on community environment. Is it friendly? Is it welcoming? Is it easy for new developers to start, and get deeper in? Are there good mentoring models? All of that makes a huge difference. And having a statement doesn’t at all guarantee anything, but it provides something people can point to and say “this is our goal.” Better than nothing, and a lot better than many open source communities are doing.
{ 3 trackbacks }
{ 9 comments… read them below or add one }
I think you meant “underserved”. I’m not sure what “undeserved” means, but it doesn’t sound like people who deserve free software ;)
Improving and promoting “diversity” (I really dislike that Eurocentric word) of open source communities may be important for the open source goals of the movement but could be hard to place in the specific project’s goals.
A NPO project in Chicago does job training and skill development around a popular open source PHP CMS. How does including ex-offenders, young single mothers or youth at risk of chronic unemployment help Project X build the best PHP CMS? It might be a convoluted line of reasoning to build that argument.
On the other hand, promoting Project X in a context with those potential users and potential developers does a lot of good. The people get access to something that can be theirs simply because they choose it. They can gain access to a community of potentially international colleagues and collaborators. And they really begin to comprehend the open source movement in the concrete and abstract.
I will be interested to see if any statements about diversity can really address the issues in a wider social context. The open source movement is a social change movement, isn’t it?
@Dustin. Uh, yeah. that’s underserved, alright. I copied and pasted that direct from their statement, so … But I fixed their typo here.
@Benjamin. If individual open source projects don’t include goals of bringing more people that the usual suspects into the project, how does this as a bigger goal of the movement make a difference? The open source movement isn’t much more than the collective communities that make it up.
I do agree that it is hard to make the argument on a project by project basis that bringing in different kinds of people makes that individual project better. And I also agree that the avenue of bringing open source projects to new groups of potential users and developers is a powerful avenue. But I can’t see how you can have a broad movement for change in open source without a bunch of individual projects grappling with the issues.
How do those users and developers from that Chicago NPO you are talking about go further in getting involved in Project X if Project X hasn’t though about, or grappled with these issues? What kind of welcome will these folks get? Possibly one that isn’t very friendly.
I don’t mean to say that individual projects cannot face the issues. I’m sorry if that’s what my comment appeared to say. I don’t want to dismiss the effort!
I’m looking for arguments that can be used in Project X’s community, and I think the case that diversity is good for the project is hard to make if we focus on the project’s own outputs. We have to appeal to values of a community, and the common elements are open source ethics, love of disruptive technology, and (vague and slippery) ideals of freedom.
Specifically about the local organization: The students have appeared from time to time at Project X Meetups, but I don’t attend regularly enough to see them again. It’s on the other side of town from me (and probably many of these students).
ok fine. i’ll join the mailing list! it’ll be the first one i’ve joined in almost 10 years!
On the more salient point Ben is bringing up — I can ask that question specifically within two communities I’m a part of.
Amanda is starved for developers. I will take anyone who offers to hack on Amanda, whether they have two legs or four. I’m sure there are things I (and Zmanda) could do to increase the attractiveness of the project to contributors, but all of our efforts have been in vain so far. So it’s hard to see how adding diversity as a goal could have much effect at this point.
Buildbot has a relatively large group of developers. Those whose real names I know have male names, so I suspect it’s completely, if not majority-male. Here, too, I would be foolish to push anyone away from the project, but there’s a case to be made for specifically reaching out to folks who I can identify as members of an underrepresented group. But this raises the larger question — aside from not lacing my slide shows with porn and using gender-neutral language, what can I do to make the project more “welcoming” to folks whose group identities I don’t even know?
I’ve realized recently that open-source projects have developed a class structure. First, consider the broad middle-class of little-known but functional projects that have a number of users and several developers (Amanda, Buildbot, Cfengine, Sphinx, Curl, the autoconf archive, maybe GMT). Above these projects are the superprojects that nearly everyone uses and that have hundreds of contributors (Firefox, Linux, Python, Perl, to name a few). These are all islands in a squalid sea of abandonware and single-developer projects (I’ll list my own ‘dev’ here, and with MM’s permission, Xena, too).
I think that approaches to diversity are wildly different for these three classes: Firefox can risk losing developers who violate project norms for inclusiveness, but projects in the middle class are “struggling to get by” and can’t afford cross words exchanged with a valuable contributor.
I believe that the intersection between non-profits (NPOs) and open source (F/LOSS) is similar to the intersection between F/LOSS and “diversity issues” (for lack of a better moniker).
What I mean is that as a consultant/developer who advocates the use of open-source software, I’m often frustrated with needing to explain in great detail all of the benefits of F/LOSS because it seems as though the concepts behind F/LOSS (most notably those vague and slippery ideals of freedom that Benjamin mentions) are often very much in alignment with or are at the very least similar enough that NPOs should be able to quickly grasp the inherent value of F/LOSS.
Similarly, as someone concerned with and cognisant of “diversity issues” (I wouldn’t deign to call myself an “activist”; perhaps “armchair activist” is the right term :), I am often frustrated at the need to have conversations in the open-source world about how to attract and welcome (or even just *not offend*) a more diverse community. It seems as though the very nature of many of the concepts that are central to F/LOSS (freedom, semi-anonymous meritocracy, value of small and large contributions, etc.) are in sync with many of concerns associated with “diversity issues”.
I’m not sure if the lessons that we many in the NP Tech community have learned from trying to explain the first situation apply to the second. I find both conversations frustratingly circular to have. I find myself saying “Don’t you see that it’s *the same concept*?!!?” and being frustrated because the similarities are so obvious to me.
So I’m not sure where else I’m going with this thought, but at the very least I’d be interested to hear of others agree that the similarities I perceive are real and if anyone has any insights on how to apply lessons learned from “one side” to the oth.
@Dustin – I think you are right on about the “class” structure of open source projects. I guess the question I would ask, is, isn’t there a middle ground? Isn’t it possible to both say a project is inclusive, and gently, but firmly call developers on their (usually sexist) comments, without alienating them to the point where they want to leave? And maybe there are women who want to work on Buildbot (or, are already working on it with a gender-neutral moniker – that happens a lot), and would feel more comfortable if there was a more open climate?
But I generally agree it’s way easier for a project like the Python project to work on this, and harder for many/most others.
One of the biggest barriers to adoption of Open Source software by nonprofits and others is the extremely poor usability. And that’s where incorporating diverse users in specific projects would really help. Women working in nonprofit orgs, at-risk youth, ex-offender groups, etc. would all bring a quite different perspective than the techie programmers to interface design and to Help Files (badly needed).
I teach a Women and Computers course which exposes students to Open Source software and I’m a somewhat knowledgeable user. It would never occur to me to volunteer for an OpenSource development project because I haven’t a clue about writing code. But I’ll bet I could be very useful in writing Manuals and making user-friendly suggestions. But invitations to join developer communities at SourceForge, etc. do not specifically or visibly invite this kind of participation.
With so many schools and universities emphasizing “service learning,” this would be a great pool for OpenSource developers to draw on.