Open Source Development: Not Very Open or Welcoming

January 21, 2010 at 9:02 pm 39 comments

We had a visitor at Georgia Tech today, alum Mike Terry, who has been studying the usability practices of open source development teams, like for Gimp, Inkscape, and Firefox.  The short answer is, “There are no usability practices,” but that’s a little too pat.  It’s a little bit more complicated than that, and actually even more concerning from an education perspective.

The folklore is that open source developers start because they have “an itch to scratch,” something that they want developed.  Mike thinks that that’s true, but that scratching that itch doesn’t actually take long. Social factors keep open source developers going — they care about their developer community and working with them.

Mike finds that few projects really care about usability.  The argument, “If you made your usability better, you’d increase your user base,” is not enticing to most open source developers.  Open source developers have no layers (like salespeople or tech support) between themselves and the public users.  Thus, they get inundated with (sometimes ill-informed and even downright stupid) bug reports and feature requests.  The last thing open source developers want is more of those.

Since open source developers soon stop being users of their own software, and they don’t want to talk to lots of users, how do they deal with usability?  Mike says that the top developers develop close relationships with a few power users, and the developers design to meet those users’ needs.  So there is some attention to usability — in terms of what high-end, power users want.

So what happens when a User Experience person wanders into the open source fold?  Mike has interviewed some of these folks (often female), and finds that they hate the experience.  One said, “I’d never have done it if I wasn’t being paid to do it.”  I guess there’s not much of an open source usability developer community.  The open source developer community is not welcoming to these “others” with different backgrounds, different goals, and most of all, not a hard-core software development background.

Mike believes that the majority of our software will be open-licensed.  I expressed concerns about that future in terms of education.

  • How do people get started in developing software in an all open-source world?  Mike suggested that open source is a great way for high school students to get started with software development.  I pointed out how unfriendly open-source development communities have been to newcomers, especially females, and how open-source development mailing lists have been described as “worse than locker rooms.”  Mike agreed with those characterizations, then said, “But once you get past that…”  Well, yeah — that’s the point.  Margolis and Fisher showed us years ago that those kinds of subtle barriers say, “This is a boys-only club — you don’t belong!” and those can prevent women and underrepresented minorities from even trying to enter the community.
  • I worry about the economics of open-source and what signals it sends to people considering the field.  Mike assured me that companies like RedHat are making money and hiring programmers — but there are many more unpaid programmers working on RedHat than paid programmers.  If the world goes mostly open source, how do we convince students that there are jobs available developing software? Many kids (and parents) already believe that software jobs are all being outsourced.  How do we convince them that there are good jobs, and they don’t have to work for years for free before they get those paying jobs?
  • Finallly, I really worry about the lack of thought-diversity in the open source communities.  People who care about usability are driven away from these communities? While we educators are trying to convince students that not all of computing is about programming, the open source community is telling newcomers that programming is all that matters.  If the whole software industry goes open source, we’re going to have a hard time selling the image of a broad field of computing.

I found Mike’s work fascinating, and well grounded in data.  I just find the world he describes a little disconcerting.  I hope that  the open source community considers the education issues of its next generation of developers.

Entry filed under: Uncategorized. Tags: , .

A National UK Animation Contest What’s worth it in reform?

39 Comments Add your own

  • 1. Tom Hoffman  |  January 21, 2010 at 11:54 pm

    I think you have to start with “compared to what?” Getting involved in say, Mozilla development might be a big leap, but it is not like getting involved in Internet Explorer development is easier. Your usability expert may have decided she’d have to be paid to work on open source. Fine… does she work on proprietary software for free either?

    Concerning the specific points:

    * I think you’re both overplaying the first point. The number of high school kids who will actually contribute to open source software projects will always be small, just because getting up to speed with a substantial codebase written by someone else is always hard, and once you start winnowing out things like projects that would actually be teaching *bad* coding habits, etc., the whole prospect starts looking sparse.

    The same is equally true, except moreso of proprietary software, of course.

    But also, I don’t know what open source communities you guys have experience with, but I don’t find them to be “worse than locker rooms.” Working your way in is inherently difficult, to be sure, but again, compared to what? Getting into your class at Georgia Tech? How easy is that?

    * The vast majority of software development does not depend on the sale of licenses. The free availability of open source infrastructure has been an enormous boon. Without open source everyone thinking of starting a web company would have to first buy a web server license, a database license, etc… Think of how much *less* economic development would have resulted from that.

    * I also think that Mike overplays the issue of open source projects actively not wanting more users and not working on usability intentionally because of that. I mean, I can think of an example of that in an open source project I use, I guess, but it is expressed as, in effect, “Fine, YOU can work on usability, but I’M not going to.”

    I would re-frame the question: How can people who care about usability create open source projects that will appeal to programmers? How can educators create a new generation of open source programmers who understand usability and its importance?

    • 2. Mark Guzdial  |  January 22, 2010 at 7:27 am

      Tom, it all depends on how you define “open.” Neither IE development nor Georgia Tech claim to be “open.” It is certainly easier to get into classes in any “Open University” in the world than it is to get involved in open source development, because the “Open University” actively and explicitly welcomes newcomers, eases the transition into the community, and works to provide entryways for a range of newcomers. Perhaps a better name for “Open Source Development” would be “Country Club Development,” because it’s actually pretty closed, not inviting, and is mostly filled with white guys. As long as the claim is “open source software development,” it’s fair to critique the movement when it’s not being open.

      • 3. Frank Krasicki  |  January 22, 2010 at 9:44 am

        Wait a minute. The open terminology simply expresses that the code is available to the well-qualified public.

        Open isn’t a social engineering movement for peace, love, and harmony. The social graces of the internet are much sharper than real life and lots of people take affront to it when there is no affront to be taken.

        And for all of the noise about UI goodness, the open-source software by leaps and bounds trumps the commercial counterparts everywhere that it can.

        I use Ubuntu on the desktop and it is simply heads and shoulders a better user experience than the infuriating Microsoft experiences my wife goes through because her school is locked into proprietary Windows-based crapware.

        If there are UI ‘experts’ claiming to have a hard time it may be that their expertise is in mimicking Windows instead of something more original. This is not uncommon at all.

        But GIMP is currently undergoing a massive UI overhaul so its hard to see where the criticism has veracity.

        But finally, the argument I hear most often in schools is not usability but that teachers have become indoctrinated to believe that the only applications business uses are Microsoft and therefore that’s what they insist on standardizing on.

        – krasicki

      • 4. Tom Hoffman  |  January 22, 2010 at 1:26 pm


        But open source development is much, much, much more inviting than any other form of development. It is the most inviting and usually the only development offering any entry into production codebases whatsoever, particularly if you are not a university student.

        And it is crucial to make the distinction between things that you would be allowed to contribute to the project and things that you can convince somebody else to do for free. The first is vastly larger than the second.

        And in the end, you can always do whatever you want with the code, even fork it if necessary. That’s what makes it “open,” not the social skills of the developers. Nobody can stop you from doing what you want, even if they don’t necessarily want to help.

        Open source development is not perfect, but are you suggesting some kind of superior alternative?

        • 5. Mark Guzdial  |  January 22, 2010 at 2:06 pm

          Hi Tom,

          I agree that open source is more inviting than commercial products — no doubt. Open source development is not perfect. I am arguing that it needs to improve. If it is going to be the dominant force in future software, it needs to consider the health of the ecosystem, which includes users’ usability and learners’ education. If it doesn’t, then it’s just a country club — just a bunch of good ole boys doing what they want, not considering the impact of locking the doors to others.

      • 6. edtechdev  |  January 22, 2010 at 11:33 pm

        “But open source development is much, much, much more inviting than any other form of development.”

        If you mean software development, then of course it is, by definition. That’s not saying much.

        “Open source development is not perfect, but are you suggesting some kind of superior alternative?”

        He’s suggested there it could be improved, and even more open, and gave other examples to show it is possible.

  • 7. Ben Werdmuller  |  January 22, 2010 at 4:20 am

    Although these are important issues for education, it’s not true that they’re specific to education. They’re hugely important to a wide range of markets, and for the success of any open source project.

    It’s telling that only 1% of open source contributors are female. These communities can be very hostile places, and very often outside voices of any kind are frowned upon – despite the “open” prefix.

    This isn’t true of all open source projects, and an increasing number will change in the future. I intend to start a new project this yer – following Elgg, the project I co-founded in 2004 – for which the evolution of the community paradigm will be equally important to pushing back software boundaries n

  • 8. Mark Guzdial  |  January 22, 2010 at 7:14 am

    Mike responded to my note over on Facebook, which I’m copying here to serve as corrections and clarifications on what I said he said:
    “Thanks for the write-up, Mark, and it was great seeing you today!

    I should make a few clarifications.

    First, there are practices in place in many projects that help improve the usability of the software being produced, but the range from ad-hoc informal practices to those steered by people with UX/usability training.

    Second, we found that open source developers most certainly *do* care about the usability of their software. However, they often focus on very specific target user groups who may not represent the majority of users (which you point out later in your note). This decision is often made to help limit the scope of the work because they are resource constrained. But, as you state, we did find that there was no real desire to increase the user base for many projects, for the reasons you state (e.g, it doesn’t lead to significant rewards for the developers).

    Note that these characterizations are mostly for volunteer-driven projects; projects undertaken mostly within a commercial framework may have some important differences.

    Third, while there has traditionally been a barrier to outsiders (UX practitioners, designers, and females), my sense is that this state of the world can, and will, change. This is a young, scrappy movement, but it is also one that is rapidly evolving. I think the interesting question is how we can facilitate positive change in the community to address these issues.

    Finally, while open source is dramatically affecting the software industry by making a lot of software packages commodities, there will always be a demand for new software to be written. But how open source will affect demand for programmers/software developers is not yet clear to me.”

  • 9. Open Source Usability « UCOSP  |  January 22, 2010 at 8:19 am

    […] education is always worth reading; today, he reports on work by Prof. Mike Terry at Waterloo on usability in open source. The good news is, there’s lots of room to […]

  • 10. Alfred Thompson  |  January 22, 2010 at 11:32 am

    In most activities one gets what is rewarded. For many people involved in open source the reward is the respect and admiration of their peers. That is not a bad goal but it does influence what people work on.

    For companies, for-profit ones at least, there is a goal to sell something and make money. That immediately means expanding the audience.This is likely the motivation, much more than altruism, behind their involvement in open source. They need the software to be more user friendly and have specific features so that they can sell the hardware to run it, the support contracts to maintain it, the training in how to use it or the consulting to install it and train users. The future of user friendly open source software is likely to remain dependent on corporate sponsorship for some time to come.

    The trick for the open source movement is going to be finding more ways that companies can make money to support the development of software that they give away. And/or to develop new motivations for people who work for free so that they have incentive to expand the user base rather than “just” show off their skills.

    • 11. Bill Fitzgerald  |  January 22, 2010 at 3:00 pm

      The differences between open source and proprietary are nowhere near as vast as some people would make out. The purpose of companies is to make money; whether or not those companies work with open source code, proprietary code, or a blend is immaterial.

      To be blunt, my dinner doesn’t care if it was paid for by work on open source or proprietary development. In my case, I work in open source, as it provides countless benefits that do not come with solutions based on proprietary code. I also believe in the open source model, and that open source development does more to eliminate social and economic injustice than other models, but that’s a philosophical position that aligns with how I run my business.

      As to the point of open source being closed, that is a disingenous argument at best, as there is no one open source community. Different projects differ widely in how they are run, and the climate will vary on a project by project basis. Like all of the technology sector, steps need to be taken to improve access. This is not unique to open source. This should not be taken as a justification of the status quo; in addressing these issues, I hope that the transparent nature of many open source communities will lead to faster, more efficient corrections of these issues. Perhaps then, these case studies can be adopted by less transparent companies, and this will provide another way that open source development/community models can help inform/change slower/less adaptive communities.

      WRT “The trick for the open source movement” — no tricks needed, really. Let’s just talk results.



  • 12. Erik Engbrecht  |  January 22, 2010 at 1:27 pm

    If you want there to be more open source projects that reflect your values and objectives then start open source projects that reflect your values and objectives.

    I think that many aspects of OSS communities represent a reaction to common corporate environments. I think corporate environments have become very culturally sterile in the name of diversity and inclusion. I think this even extends to the way different technologies and skills are treated. There’s also the classic problem of the politically adept being promoted over the technically adept. And finally, and probably most importantly, is that commercial work is often boring. Even interesting projects have large expanses of boring parts. I’ve heard others comment that the more interested you are in your day job, the less likely you are to dedicate professional energy towards open source. I think for some contributing to open source is a dream that someday someone will come along a sponsor their contributions, so they can leave boring day jobs.

    Open source can be an escape from project managers, business analysts, QA people, configuration managers, incoherent customers, iron-fisted political correctness, user interface experts, and countless other measures in place to ensure that a usable product is produced in a manner that doesn’t yield lawsuits from employees.

    I think there’s a only one way to actively steer a project towards a specific culture: be a leader on that project. You do it from the inside by working your way up or from the outside by starting projects.

    I think all the commentary on how OSS projects are hostile just serves to perpetuate it. It creates an us versus them divide, and it encourages people who might be able to find a friendly project to not even try by over generalizing behavior.

  • 13. Michael Kölling  |  January 22, 2010 at 1:48 pm

    I very much agree with all the main points in Mark’s post. And I think the point about the rewards for open source developers is related to the usability problem.

    Open source developers get their rewards for their work by getting recognition from their peer group. That recognition usually stems from implementing new, useful features in a project.

    Features, however, often conflict with usability. Adding more features makes it more difficult to create a good user experience, especially for beginners and casual users. And the complexity is not linear: When adding more features, the user experience complexity rises somewhat exponentially (in my completely scientifically estimated experience).

    To achieve a good user experience for average users, there is often a point where new features should be rejected. If the software is already complex, features may have to be removed to make an improvement.

    However, there is no peer recognition in open source communities for NOT doing something. The only way in and forward for programmers to earn their reward is via adding features. It is a necessary result of design by committee that a system will grow to a state of complexity that is appropriate only for power users.

    The point – made in a comment above – about Ubuntu being user friendly does not contradict this. Ubuntu only came about after professional usability experts were brought in by a centrally controlled organisation. Linux as a distributed, open, peer-driven effort did not manage to achieve acceptable usability for casual users for many many years before that.

    I also especially agree with the problem Mark points out that this culture is highly discouraging to women entering these projects. This is a serious problem that worries me even more than the usability implications. We have to find a way to create a software development culture that does not a priory alienate half the population.

    – Michael

  • 14. Tom Hoffman  |  January 22, 2010 at 3:07 pm

    The thing is, there is no “it.” There is no “other thing” that is open source development — unless you define “it” as unwelcoming white male volunteer nerds. Then, by definition, open source communities are problematic.

    It is not less “open source” when someone is being paid. Michael writes “open source developers get their rewards for their work by getting recognition from their peer group” and “this culture is highly discouraging to women entering these projects.” Well, does that apply to him? To his open source projects? Or are his open source projects not really open source projects because they aren’t sexist, peer-recognition driven, etc.?

  • 15. Bell  |  January 22, 2010 at 7:26 pm

    The description of the good ole boys locking the doors on all and sundry raises my hackles in all sorts of ways. For how bad it would be if it were correct and how incorrect it is.

    There are usability problems with open source software, and the reasons are not going to be easy to address. It’s got nothing to do with the community being (to quote a horrid safferism) “pale, male and stale”.

    The first, and most difficult to address reason is that user interaction is verbose to write, laborious to test and painstaking to debug. For web front ends there are some aids in this regard, but they don’t completely remove the pain. For the contributors who are there for fun, rather than being commercial programmers working on an external code base, this lacks appeal in spades.

    Which brings us to what motivates volunteer coders, which is important because these are the people who ultimately have to do all the bits. Many are hanging around in projects after scratching their itches. Older projects have had people graduate from interested users to developers just by momentum. This may be the best case for wanting to gain peer recognition, for the people you looked to for answers at the start to acknowledge your progress. Beyond this the case for peer recognition powering the development effort weakens a bit, replaced by personal feelings of achievement or a need to be a responsible FLOSSizen.

    Programming is not all that matters. Pushing an agenda, usability or any other, without being able to complete the items yourself comes across as managing though. This doesn’t make friends in small, peer directed teams. It doesn’t help that usability practitioners, particularly those newer to the field who see usability as a human right rather than a tradeoff, can get a bit shrill.

    As projects mature and teams get larger a lot starts to change. Once a team feels that they their project has reached a level of completeness (often based on being able to address a generic enough case of the itch that started the scratch) it’s time to go looking for a bigger user community. From the bigger user community will come a bigger development community. Easing access to this development community becomes important and some projects rise admirably to the challenge. Postgres, for instance, has great mailing lists and a fantastic set of documents and utilities which come with the source code download to make getting up to speed as a new developer quicker and easier. These developer intro tools, like test datasets, are an ideal place for student involvement in a project.

    • 16. edtechdev  |  January 22, 2010 at 11:44 pm

      “The first, and most difficult to address reason is that user interaction is verbose to write, laborious to test and painstaking to debug.”

      That’s actually a great practical point. To do usability testing currently requires tens of thousands of dollars worth of proprietary software and equipment (Tobii, Morae, and the like). Also, there are numerous books, but no really thorough free, openly licensed resources for learning much about usability testing.

      Would’t it be great if some funding/sponsorship targeted the development of full-featured and open source / open hardware usability tools and training materials.

      There have been small projects here and there, but still a long way from the level of Tobii, Morae, and the like.

      • 17. Mark Guzdial  |  January 23, 2010 at 8:15 am

        Mike Terry has been running workshops at Open Source Developer conferences showing how to usability testing and development on-the-cheap. It’s not hard to get better than the current state in most projects.

  • 18. Mark Guzdial  |  January 22, 2010 at 9:12 pm

    For those interested in some additional empirical data about the lack of diversity in Open Source Software Developmen (as Tom put it “as unwelcoming white male volunteer nerds”)t:

  • 19. Tom Hoffman  |  January 23, 2010 at 12:44 am

    Mark’s link above cites the FLOSSpols study the EU did on the subject of women’s participation in open source. And it raises some important points, and things open source projects need to do better. The difference between their findings and Mark’s attitude, as I read it, is that there is no hesitation in the EU findings about the idea that the key to fixing gender inequality is getting more women involved in open source projects. That from a policy point of view there needs to be push, including from CS departments, and pull (from the projects themselves).

    • 20. Mark Guzdial  |  January 23, 2010 at 8:16 am

      I agree with you and the EU study, Tom. Getting more women involved would be a great thing, with push-and-pull just as you describe. How to achieve that pull is an interesting question, and the willingness to make that pull may lead to more openness overall — curb cuts help everyone.

  • 21. Tom Hoffman  |  January 23, 2010 at 10:47 am

    I’ve kinda reframed the issue on my blog:

  • 22. James Howison  |  January 23, 2010 at 5:13 pm

    One of the tricks of usability in Open Source is that achieving it usually requires direct interdependent collaboration between two (or more) people. And that is hard to achieve in any distributed volunteer situation.

    In my research work [1] I’ve found that the vast majority of work in open source projects is accomplished in short bursts by one individual doing the coding; if I’m right about that then collaboration in the way that is envisaged in the traditional usability paradigm (multiple rounds of dependent interaction) is a problem built into the way that open source projects organize.

    Mike’s work is really interesting because it points out why open source projects don’t necessarily want to attract more users.

    The way I see it usability requires thinking right across a codebase, so it has many more interdependencies than, say, adding a feature. In short you have to alter the behavior of multiple people—people over whom you have little control—and do it in ways that are fundamentally opposed to self-motivated volunteer work.


  • […] people contacted me after my post about open source to tell me how their open source project cares about usability, or has lots of women on it.  I am […]

  • 24. Chani  |  January 25, 2010 at 12:22 am

    it’s not all doom and gloom. KDE is a large, diverse free software community that cares about usability and welcomes everyone, regardless of gender or cultural background or whatever. we’re not perfect, but more and more projects within KDE turn to our usability people for advice when adding UI, and there’s hardly any “locker-room” behaviour – personally I’ve only been mildly irritated by a single developer, and had to /ignore one obnoxious user on irc several years ago.

    I’m not sure what it’s like outside of KDE… I hope it’s not really as bad as you describe. personally I think I’ll just stay in this happy kde world (and get a job in it too – companies seem desperate for skilled Qt developers these days)

  • 25. Ben Lafreniere  |  January 25, 2010 at 10:12 pm

    Regarding the “country club” feel of OSS projects, there was a very good paper at CSCW 2005 by Nicolas Ducheneaut titled “Socialization in an Open Source Software Community: A Socio-Technical Analysis”. He looks at the complex process that developers go through to successfully join an OSS project and gain permission to make large changes to it. The process is one of slowly building up social credit with the community over a long period of time, subtly influencing the code through patches, and learning the—often unwritten—social structure of the community.

    Pertinent to this conversation, he makes the point that OSS projects are set up in this way to prevent large disruptive changes from the outside. That is, it’s a coping strategy in lieu of things like formal hiring procedures for new developers in closed software development environments.

    If he’s right, then this is hopeful. It means that OSS projects can be more welcoming to outsiders joining the project, provided that other strategies are put in place to protect against large disruptive changes. The Mozilla foundation, Canonical, and Drupal may all be examples of this in action. They have all imposed further structure on their projects, and brought in people to address usability issues.

    • 26. Tom Hoffman  |  January 25, 2010 at 10:25 pm

      Part of how you perceive this issue overall is whether you think that the aggregate open source community is evaluated by the thousands of little projects, with big projects like Mozilla representing an outlier, or whether you consider the core of open source to be Ubuntu, Mozilla, Drupal,, etc., with a long tail of tiny projects of little significance in themselves.

      • 27. Ben Lafreniere  |  January 25, 2010 at 11:01 pm

        That’s an interesting idea. To expand on it, I think that there are routes for inclusion in both of the extreme cases.

        The large, iconic projects appear to develop their own organizational structures to guide their development, and with that comes an openness to usability issues and more inclusiveness in their community of contributors (if we take Drupal, Mozilla, and Canonical as examples).

        Very small projects, on the other hand, could conceivably be forked if the developer community is toxic to certain groups, or ideas.

        So that might leave an awkward middle, where projects are too big to easily fork with good effect, but still have a primarily ad-hoc culture, and with it an environment that makes it difficult for outsiders to enter.

  • 28. Alfred Thompson  |  January 26, 2010 at 11:22 am

    I am very much the outsider in the the open source discussion. I have spent the better part of my 35+ year career working for companies that make and sell proprietary software. While I have worked on projects that more or less follow the open source model of development these projects have been pretty small and not in the LINUX world at all.

    So how do I judge the welcoming ways of the open source world? By the way the people who self identify with it interact with me. Those experiences have been overwhelmingly negative. It is as if they see me as irredeemably tainted and the enemy of all that is good and right in software. This is of course not a good way to win me over. 🙂

    People in this thread see open source as more welcoming of people than proprietary software projects. I had one big dream after college and that was to one day be a developer on the RSTS/E (a proprietary) operating system. I did get the chance and it wasn’t all that hard for me. Dang, I am batting 1 for 1. Home run the first time. So my personal experience has been that people are much more welcoming in proprietary software. Send in a resume, get an interview, people offer to pay you money to write code. How is that not welcoming?

    What is the model for open source? It appears (remember I am on the outside looking in) that one can get the code and start working but that it often takes quite a long time and lots of hard work before people are willing to take and use your code. And then they still take it for free. You get a pat on the head and people use your code and that is about it. This is encouraging to people who need to make a living? Really?

    It seems as though the process to work on open source as a company employee is pretty much the same as for working on proprietary software though. In any job having a portfolio of work to show people is an edge – open or proprietary. And frankly my observation is that makers of proprietary software are way a head in realizing the importance of a diverse population of contributors. They are well out in front trying to recruit women, minorities and all sorts of different people into the development process. Diversity is a core value where I work. Is it in open source? I don’t see it.

    • 29. Erik Engbrecht  |  January 26, 2010 at 12:11 pm

      I think you make a very interesting observation:
      “So how do I judge the welcoming ways of the open source world? By the way the people who self identify with it interact with me. Those experiences have been overwhelmingly negative. It is as if they see me as irredeemably tainted and the enemy of all that is good and right in software. This is of course not a good way to win me over.”

      Last night I was pondering exactly this as a kind of thought-experiment, but I lack the background to have experienced it.

      Here you are, a privileged white male, and you experienced open hostility in the open source world due to differences in your technical background and/or preferences…and possibly a resoundingly negative view towards the production of software without direct compensation.

      The more I think about it, the more I think project outsiders tend to have a very different view of a project’s needs than project insiders.

  • 30. Bradley M. Kuhn  |  February 17, 2010 at 6:04 pm

    Mark, I think that you perhaps were not aware of how much interest and desire there is about usability in FLOSS, and also how often FLOSS volunteer involvement leads to real employment for many developers. Finally, the problems of diversity and sexism are serious ones, but I believe they are shared throughout all of computing (including academic CS), not just in FLOSS specifically. I have written a complete blog post discussing these points at length, which complements Greg DeKoenigsberg’s comments.

    • 31. Mark Guzdial  |  February 17, 2010 at 9:14 pm

      Bradley, according to NCWIT only about 24% of all IT jobs are held by women. Your blog post suggests that 22% of faculty are women. According to the FLOSSPOLS study, only about 1.5% of open source developers are women. That’s a much larger problem in the FLOSS community. To respond to the issue in the commercial and academic communities, BPC and NCWIT were established. What is the FLOSS community doing at a similar level of seriousness to address this problem?

      • 32. Bill Fitzgerald  |  February 17, 2010 at 9:55 pm

        Mark — there is one important point to be drawn from your post: sexism continues to exist in the tech world, and that reality deserves serious attention.

        But the place where your argument rapidly deteriorates is when you attempt to create one homogeneous “open source” world, with a single ossified worldview. The ecosystem of people that participate in FLOSS development (code, docs, usability, testing, etc) is incredibly broad, and each of these communities has its own rules, and norms. There are numerous (dare I say countless?) FLOSS projects, but pretending that all of the FLOSS projects strewn about what some people call the “FLOSS community” actually listen/respond to a central authority is laughable. Initiatives are undertaken on a project by project basis.

        If you are interested in what success looks like, look at what is going on in projects where there is active, successful recruitment of new members. But this is more difficult, as it requires spending time working within the community, and I have yet to see any decent academic work on how open source communities thrive. On some levels, its more of an anthropological question than a computer science question.

  • […] to recent discussions here about Blacks, Latinos, and Women in Silicon Valley and the non-open culture of open source software development. Unfortunately, computer geeks are generally really oblivious to their class privilege, mistaking […]

  • 34. msk  |  February 28, 2010 at 4:01 pm

    All I can say here, one has to take a look at what our institute (and other institutions) have tried to address – with a generous donation from a distinguished alumni, we (RPI) are off to a great start. Please take a look at Rensselaer Center for Open Source Software and Dashoboard of projects and the links there in. I am not discounting some of the difficulties we still face. How do we sustain our efforts over a long period of time? How doe we make our administration realize that this is a worthwhile effort? These are the challenges that every one faces – This does not make open source more closed – Answer to any “open” question is “closed” after only one google query 🙂

  • […] the book anyway on his own?  I worry about insuring quality for millions of schoolchildren when the track record of most open-source software projects is that they tend to care about the usability…, rather than making sure that everyone can use it.  That won’t work with […]

  • […] or as a way to gain experience with significant code bases.  I’ve mentioned before that the OSS community doesn’t have a great track record for diversity and welcoming newcomers. Here’s a new study describing how hard it is for newcomers to connect with the oldtimers in […]

  • […] An interesting new piece on identity within the open source community.  Noah Slater addresses a concern that I have, that the definition of contribution in open source communities limits the opportunity for legitimate peripheral participation. […]

  • […] to write the docs or to do user testing is not a form of legitimate peripheral participation because most open source projects don’t care about either of those. The activity is not […]

  • […] * Mark Guzdial : Open Source Development : Not Very Open or Welcoming […]


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Trackback this post  |  Subscribe to the comments via RSS Feed

Enter your email address to follow this blog and receive notifications of new posts by email.

Join 10,184 other subscribers


Recent Posts

Blog Stats

  • 2,054,518 hits
January 2010

CS Teaching Tips

%d bloggers like this: