Log in / Register Account

It is a convenient myth for a lot of people in the free and open source software community that our projects have few barriers to entry beyond a base set of knowledge about the project new contributors want to try to join, and the skills need to contribute to a project.

Diversity, to a lot of people who buy into the pure meritocracy myth, is a problem that can be solved by accepting anyone who can contribute. It’s the contribution that matters, not the person’s race, gender, or other identifiable status. Train more people up, the meritocrats will argue, and the diversity problem will be solved.

In reality, this is not usually the case. And it’s not just for the obvious (and very real) issues of boorish white cis men who actively seek to empower themselves by belittling those who are not like them. I’m not dismissing these concerns, and I believe every community needs to work to counter such negative behaviors.

But the problem goes much deeper than altering or removing the behavior of bad actors.

During Sunday's keynote at DevConf.us, Red Hat CTO Chris Wright called out another major obstacle to diversity: the groups we are seeking to include are also experiencing systemic barriers to entry. It’s all well and good to promote the learning of programming skills across many diverse groups, but what a lot of technologists may not comprehend (because it’s never been a problem for them) is that the resources to learn about such contributing (such as a computer or stable Internet connection) may not be readily available.

There are, of course, programs and schools in place to offer folks these tools and skills. And these are certainly great. But there is also the issue of time, something else a lot of us take for granted. If you have school and maybe a part-time job (or two); a job or jobs that take you past a 40-hour week; or all of these and a family to care for--these scenarios could make the time needed to learn technology skills a very scarce commodity.

Socioeconomic status, Wright told his audience, can raise huge barriers for the very people we as open source communities want to bring into the fold as contributors. This is why, when we do see new contributions coming in from such individuals, we need to value those contributions and ideally nurture the contributors’ efforts, for the simple reason they may have had to put in far more effort just to get to that point than individuals who represent a less diverse socioeconomic or ethnic group.

Increasingly, however, the efforts of many individual contributors may be danger of being devalued altogether.

Paid vs. Volunteer

Overarching the problem of attracting a diverse set of contributors to open source projects is the increasing observation that volunteer contributions may be getting lost or less valued than contributions coming from paid developers who jobs enable them to work on a given project or projects.

There are known cases of “outside” contributions being increasingly ignored or rejected by corporate developers whose company “owns” and open source project. Sometimes this may happen for a reasonable technical need. But sometimes it happens because a shepherding company doesn’t want to go in the direction of the contribution for business-oriented concerns.

Never mind the fact that the contributor may have really needed that change. Or the change may have led to a new avenue for innovation.

It isn’t even just a problem of active barriers. If you have a volunteer who can only contribute 2-3 hours a week on a project, because that’s all the time they have, there is a chance that just by signal-to-noise ratio, those contributions will be washed out in the deluge of contributions coming from the full-time paid contributors to the project. Or made obsolete when the project takes a sharp (and unannounced) turn in the time between the volunteer’s contributions.

Open source projects with corporate participants must take great care not to ignore the efforts of anyone who contributes infrequently. These efforts must be valued on technical strengths, not frequency of contribution.

Failure to do so will not only close avenues for inclusion of diverse contributors, but will also close off new and potentially innovative contributions from any contributors not on the company payroll. That will be a very lackluster environment for any volunteers to participate, and will ultimately lead to communities dominated by corporate interests alone, not diverse talents and points of view.

Image by geralt, under CC0 1.0 Public Domain Dedication.

About the author

Brian Proffitt is a Manager within Red Hat's Open Source Program Office, focusing on content generation, community metrics, and special projects. Brian's experience with community management includes knowledge of community onboarding, community health, and business alignment. Prior to joining Red Hat in 2014, he was a technology journalist with a focus on Linux and open source, and the author of 22 consumer technology books.