How Men and Women Approach Open Source Differently
Squashing Inclusivity Bugs in Open Source: Dr. Anita Sarma Shares Her Research
With open source software becoming more and more a requirement for job searchers, job seekers are finding that it is critical to becoming a part of this community. Many would rather see a GitHub profile than a CV.
But currently, only about 10 percent of open source contributors are women.
For this episode of The New Stack Makers podcast, Dr. Anita Sarma, associate professor of computer science in the Department of Electrical Engineering and Computer Science at Oregon State University, joins us to talk about her recent research on how to increase gender inclusivity in open source.
Dr. Anita Sarma, Oregon State University
Her recent research focuses on problem-solving facets in which men and women differ statistically. In her research, she has focused on five ways in which men and women statistically differ in how they problem solve.
By understanding these different problem-solving facets, as she calls them, we can begin to understand how to become more inclusive. Which leads to better software, and increased profits.
Caveat: these are broad strokes and not all men and not all women break so easily into these categories, So, not all men and not all women, but there is enough of a statistical difference to, well, make a difference.
For example, men and women differ significantly in their learning styles. Women tend to be process oriented, learning step-by-step. She described men as “tinkerers,” clicking here and clicking there, exploring in a round-about way to find what they are looking for.
One style is not better than the other, but open source should accommodate both approaches. So if most Open Source projects do not have documentation with steps on what the software does, the people who have learned in a process-oriented way will have a much harder time contributing to that project.
Next is what Sarma calls the “confidence-competence gap.” Women, on the whole, have far less confidence in their work than men at the same level. What this means for open source is that if something goes wrong, more women than men blame themselves, which often leads to them quitting.
More women than men try to learn about technology because it’s part of their job. Men want to learn about it because it’s cool.
The last facet she’s researched is the attitude towards risk. Just like more men than women are tinkerers, they are also less averse to trying new things as a whole. “Oh let me try this new feature, oh, let me update this operating system, what can go wrong,” said Sarma. More women might say, “Oh the deadline is coming up, let me work on what I know, even if it is not the newest feature.”
Not one to be interested in pure academic research, Sarma has some suggestions for small things that can be done to improve the inclusivity of open source projects.
How to Make Your Project More Inclusive?
Sarma suggests using the GenderMag approach to teaching, from the Human Interaction Project. While they weren’t developed for open source software, applying the principles to an open source project can surface areas in which the project can improve its accessibility. By creating a persona of the types of persons generally excluded and doing a cognitive walkthrough of the project, areas of exclusion can surface.
Listen to hear a deeper dive on issues surrounding open source documentation, why pre-code contributions are critically needed, what a “push-ocracy” is and some specific things you can do to make your project more inclusive, and thus increase its chances of success.
In this Edition:
2:55: Cognitive styles and statistics between individual genders, information processing, and problem-solving.
5:23: Why open source should be more inclusive based on research done from other organizations.
10:01: Documentation, open source, and “pre-code contributions.”
17:46: Exploring the GenderMag tool, from the Human Interaction Project.
21:39: Who exactly is using this tool?
25:11: Confidence versus competence.
Raygun sponsored this post and podcast, which were produced independently by The New Stack.
Feature image via Pixabay.
